# Changelog
All notable changes to vtcode will be documented in this file.
## v0.73.2 - 2026-01-29
## 0.136.2 - 2026-07-19
### Highlights
#### Bug Fixes
- Only exclude top-level prompts/ from crate tarball (6439c576)
### Other Changes
## 0.136.1 - 2026-07-19
### Highlights
#### Bug Fixes
- Shorten invalid crates.io keyword (1a7e5c9c)
- Skip missing embedded assets during publish verification (317f9430)
### Other Changes
## 0.136.0 - 2026-07-19
### Highlights
#### Bug Fixes
- Replace private intra-doc link with literal value (1a643383) (@vinhnx)
- Remove jemalloc feature from Windows builds; format release.sh (48452837) (@vinhnx)
- Prevent gh run view from opening interactive pager on CI failure (cb208c11) (@vinhnx)
- Resolve 3 failing tests after workspace reorganization (438119cb)
- Robustly detect workspace root and propagate CLI override errors (c897aee7)
- Resolve link_section warnings and unblock vtcode-core build (aa8a4b57)
- Resolve clippy warnings and dead code across registry and compaction (873e5473)
- Handle blocked planning tools by ending turn gracefully (e44794ed)
#### Documentation
- Add TODO for API key UI/UX and auto-discovery (c118c464)
#### Features
- Warn if trusted plugin directory is writable by group/others (cff365e7)
- Implement permission events and preflight flow for policy-denied tools (f15cd7df)
- Improve credential discovery with env var detail and paste UI (a6dd9480)
- Add /secret slash command for managing secrets (eb279628)
### Other Changes
#### Other
- Update README.md (967799ac) (@vinhnx)
- Update README.md (e869ec35) (@vinhnx)
- Update README.md (a6105983) (@vinhnx)
- Update TODO (6ff1ed1d)
- Update TODO (1910ff03)
- Update TODO (0165ff68)
#### Refactors
- Restructure extensions and xtask locations (20367519)
- Decompose file palette monolith, fix last_entered fragility, add tree-style rendering (413ccd2e)
- Remove unused dependencies and improve safety documentation (6a7d47dc)
- Rename session-store to memory, add harness_emitter (28b451f0)
## 0.135.14 - 2026-07-18
### Highlights
#### Bug Fixes
- Revert setup-zig to upstream and remove --frozen from builds (07b226ea)
### Other Changes
## 0.135.13 - 2026-07-18
### Highlights
#### Bug Fixes
- Add workspace dependency versions for crates.io publishing (ef854c98) (@vinhnx)
### Other Changes
#### CI
- Use local setup-zig action with node24 (db171faa) (@vinhnx)
- Optimize platform build bottlenecks across CI and release (e1612291)
#### Other
- Update project (32eaa76e)
- Update TODO (b2540992)
#### Refactors
- Simplify auto-permission prompts and asset sync, remove redundant system prompts (c625737e)
## 0.135.12 - 2026-07-17
### Highlights
#### Bug Fixes
- Resolve clippy and compilation issues for publishing (d8129022)
#### Documentation
- Add config fields for system prompt token budget and trimming (a10a65ac)
- Update TODO.md with Phase 5 verification and remove diagnosis section (92144b5b)
- Add battery-packs blog post and register in index (977dd194)
- Update extension docs for MIT OR Apache-2.0 license (522be78d)
#### Features
- Add vtcode-battery-pack crate (121f4bde)
- Add startup config warnings and measure_system_prompt_size (3e64fe14)
- Add Apache-2.0 dual license and Codex attribution (876e4f63)
- Generate THIRD-PARTY-NOTICES and fix attribution refs (0515114d)
- Restructure to crates/{common,codegen}/ layout and upgrade toolchain (c217c68c)
### Other Changes
#### Other
- Revert "docs: add battery-packs blog post and register in index" (6a0eafb9)
- Revert "feat: add vtcode-battery-pack crate" (7b2fdabd)
#### Performance
- Bounded channels, async offload, and runtime backpressure (5271831b)
## 0.135.11 - 2026-07-17
### Highlights
### Other Changes
## 0.135.10 - 2026-07-17
### Highlights
#### Features
- add Moonshot models, update model presets, fix blocks reflow (95ae2bcf3) (@vinhnx)
- show visual TUI feedback for every /config change (eb42b19fe) (@vinhnx)
#### Bug Fixes
- remove trailing empty Policy line from streaming reasoning blocks (be920c5a0) (@vinhnx)
- restore PTY/command tool-call display and register exec_pty_cmd (225c5c1ae) (@vinhnx)
- persist salvaged plan to session file during plan-mode recovery (c5dcc9f6a) (@vinhnx)
#### Documentation
- note plan-mode investigation in TODO (92298523b) (@vinhnx)
### Other Changes
#### Performance
- share conversation history via Arc to eliminate per-turn O(history) clones (7faf732c5) (@vinhnx)
- cut blocking IO and redundant work on tool/event hot paths (cb012b255) (@vinhnx)
- cut per-turn and streaming overhead across hot paths (e310ec8cb) (@vinhnx)
- eliminate repeated work in search hot paths (669d5b9e8) (@vinhnx)
#### Refactors
- extract ColorContext, simplify thinking header, fix PTY completion line (124629ffd) (@vinhnx)
- consolidate PTY styles, deduplicate prefix-building, relativize paths in headers (23c75549d) (@vinhnx)
- harden public/harness tool dispatch and bump Rig to 0.40 (4619b1cb2) (@vinhnx)
- lazy flat file palette with on-demand listing (8ca10317c) (@vinhnx)
- harden native_plugin FFI and document field lifetimes (7c5a77bcd) (@vinhnx)
#### Other
- Update TODO (b6274d124) (@vinhnx)
- Update note (e7b5878e8) (@vinhnx)
- Refine plan mode (4a7a76f21) (@vinhnx)
- Update TODO (cfc90d6b0) (@vinhnx)
- Feat/simplify search (#702) (d7232ef41) (@2789734+kernitus)
### Contributors
@vinhnx, @2789734+kernitus
## 0.135.9 - 2026-07-14
### Highlights
#### Bug Fixes
- Add missing cfg(unix) attribute for env_lock import (ff652e62) (@vinhnx)
### Other Changes
#### Other
- Refactor/tools (b5e2b34f) (@kernitus)
- Fix plan mode (4ceb862d)
#### Refactors
- Unify plan-mode phrases and decompose workflow monolith (a505dd77)
## 0.135.8 - 2026-07-14
### Highlights
#### Bug Fixes
- Remove unused vtcode-core dep from vtcode-eval so it can publish (0f3b92dd)
### Other Changes
## 0.135.7 - 2026-07-14
### Highlights
#### Bug Fixes
- Add version specifications for dependencies in Cargo.toml (b2f00d1e)
### Other Changes
## 0.135.6 - 2026-07-14
### Highlights
#### Bug Fixes
- Finalize plan on wall-clock exhaustion instead of interview loop (2493ecd7)
- Strip tools at API level after wall-clock exhaustion (1a935c8d)
- Enhance plan-mode fallback to reject non-plan salvage (a329fe38)
- Trim whitespace in plan-mode recovery fallback to prevent garbage injection (f9e4d32e)
#### Features
- Implement budget exhaustion handling in planning workflow (789845e9)
- Implement user approval for entering planning workflow and enhance recovery directives (1eefd21b)
- Enhance planning workflow with agent prompts and intent phrases (19e60844)
- Add thinking block collapse functionality and default state (76d30f9a)
- Improve thinking block toggle and reflow cache handling (0434e31b)
- Enhance thinking block rendering and state management (930ca639)
### Other Changes
#### Other
- Implement primary agent switching in planning workflow and enhance error handling with retries (53bfcfe3)
- Add recovery handling for planning synthesis truncation and enhance session state (623e88db)
- Update default configurations for tool result clearing and client tool search; enhance token efficiency warnings and documentation. (3b66f9f7)
- Implement planning workflow execution and intent detection modules (6ffa4db8)
- Add tests for agent runner configuration and tool exposure (23d2fedf)
#### Refactors
- Streamline plan mode recovery and enhance user-facing messages (dd48dc0a)
## 0.135.5 - 2026-07-12
### Highlights
### Other Changes
#### Other
- Add evaluation suite command and context reset functionality (53beae8d)
- Add eval suite runner and context reset logic (5ecc0441)
## 0.135.4 - 2026-07-12
### Highlights
### Other Changes
#### Other
- Implement progress monitoring and reinforcement learning engine with persistence (233f9751)
- Refactor context engineering documentation and enhance planning recovery in turn loop (c895b2c7)
- Refactor post-tool recovery logic and enhance planning recovery handling (bd0aed15)
- Refactor read extent handling and improve recovery fallback logic (bbab7878)
- Implement tool result reducers, evaluation rendering, and tool rejection handling (c17728b3)
- Update TODO (9924d22d)
- Refactor assertions for floating-point comparisons and improve case-insensitive path checks (9152aae4)
## 0.135.3 - 2026-07-11
### Highlights
#### Bug Fixes
- Namespace cache per target in CI to prevent collisions (014e8924)
#### Features
- Add /checkup command and optimize diagnostics output (7d26d11a)
- Enable debug assertions and overflow checks in release builds (f8768d32)
- Implement mode switch guard to prevent cycling during active turns (7551e9bc)
### Other Changes
#### Other
- Implement guards for tool call validation including read, shell run, spool, and task tracker guards (e9e19936)
- Add Claude Advisor server-side tool configuration and integration (7612fb8a)
- Implement response chain handling, turn snapshot capture, tool shaping, request gap tracking, and usage cost estimation (73b1019f)
- Implement automatic compaction orchestration and memory envelope management (2743ca18)
- Add unified per-session state store with append-only event log and derived views (9d6901cd)
- Implement unified executors for cron, mcp, and agent actions; enhance tool descriptions and parameters; improve output processing for large results. (22276fb4)
- Consolidate agent and MCP tool registrations; update parameters and descriptions for clarity. Enhance action dispatching for wait and close actions in agents. Adjust risk scoring for MCP connect/disconnect actions. Remove unused parameter schemas and streamline collaboration specs. (69222268)
- Add tests for memory lock functionality and implement atomic file writes (f5055f80)
#### Refactors
- Expose normalize_model_id function and simplify model normalization logic (6efca878)
## 0.135.2 - 2026-07-10
### Highlights
### Other Changes
## 0.135.1 - 2026-07-10
### Highlights
#### Bug Fixes
- Use u64 for assumed system memory constant to fix 32-bit builds (4db0d8d6)
#### Features
- Implement read limits configuration and update related cache handling (0f47cb4b)
- Implement session mode resolution and persistence logic (be087eae)
### Other Changes
## 0.135.0 - 2026-07-09
### Highlights
#### Bug Fixes
- Comprehensive code review and debugging fixes (060baf81)
- Skip CARGO_INCREMENTAL with sccache wrapper in check-dev (b3009d4a)
- Address critical/high/medium bugs and DRY violations across codebase (099e7d60)
- Dedup recovery dispatch, fix error swallowing, clean DRY (e403b78a)
#### Features
- Differentiate model-switch outcome variants, add harness opt-out config (b535d70f)
- Add immediate submit commands for slash palette (aa01bce9)
- Include enum options in preflight schema hint (74a9ca18)
- Add support for GPT-5.6 models and update related prompts and configurations (6775385c)
### Other Changes
#### Other
- Add planning intent detection and validation enhancements (8176201f)
- Update TODO (c7b353de)
#### Performance
- Lowercase once per strip instead of per marker scan (83a33aaf)
- Pre-size summary prompt and retained-message vectors (b8de575b)
- Pre-size preflight validation failures vector (95ba7560)
- Pre-size system-prompt line vectors (4d583287)
- Pre-size render output lines and list stack (35a90e05)
- Pre-size render output strings (83407ee2)
#### Refactors
- Name failure mode in recovery retry directive (593b4c67)
- Extract finalize_turn helper from run_turn_loop (85460ba6)
- Improve error handling and streamline post-tool recovery logic (bb3aeff2)
#### Tests
- Cover tool-free recovery retry-then-salvage (00f6ad3f)
## 0.134.15 - 2026-07-09
### Highlights
#### Bug Fixes
- Primary agent runtime config clobbers configured model with "inherit" (fe45c4ec) (@ForrestThump)
- Retry tool-free recovery on contract violation instead of bailing (ebc936ad)
- Raise recovery synthesis token cap 1024 -> 4096 (383c7151)
- Update contributor commit counts in README and add WIP note in TODO (891da9e0)
- Stop injecting contradictory resume directive during tool-free recovery (7acc042d)
#### Features
- Implement absolute line cap for read_file calls with continuation support (a345212b)
- Add absolute line cap for line-based read_file operations (c2d7a50f)
- Implement WorktreeReconciler with DiffVerifier for safe merges (58c257c1)
- Salvage prose from rejected recovery synthesis in fallback (c30ac37d)
- Enhance local model management and readiness checks (f7271b25)
### Other Changes
#### Build
- Eliminate last 2 unsafe blocks in build.rs and process_hardening.rs (650781c6)
#### CI
- Wire allocator-jemalloc feature into build-linux-windows.yml (e5f4b3d6)
#### Other
- Remove sccache + rust-cache from Windows and macOS builds (ecd69fc1) (@vinhnx)
- Update TODO.md (51f4dad0) (@vinhnx)
- Delete PLAN-loop-engineering.md (40ae26f0) (@vinhnx)
- Delete web-fetch-hilt-fix.md (05425cda) (@vinhnx)
- Update TODO.md (e8266c24) (@vinhnx)
- Update TODOs (66a93a98)
- Update TODO (ec8e24ee)
#### Performance
- Shallow git checkouts in CI, enable sccache for local macOS release builds (d55f76ec)
- Default to jemalloc on Linux and add bench-allocator RSS diagnostic (998f101e)
## 0.134.14 - 2026-07-07
### Highlights
#### Bug Fixes
- Resolve warnings and improve CI status reporting (cff3a5b7) (@vinhnx)
- Improve CI status reporting to reflect accurate build conclusions (c3077862) (@vinhnx)
- Fix #693: enhance model resolution and testing for local and custom providers (369cf8bc)
#### Features
- Add nord and nord-light themes with syntax highlighting support (2385d63c)
- Update theme colors for improved WCAG AA contrast compliance (cddb8f1b)
- Enhance diff rendering styles for better accessibility and theme support (50152ddf)
- Increase codegen units for improved parallelism during check/clippy (9d57197a)
### Other Changes
#### CI
- Add rust-cache + sccache to all build jobs; use --cargo-profile ci for nextest (016ad679) (@vinhnx)
- Remove redundant check job; add SCCACHE_CACHE_SIZE; consolidate sccache env (d37c812c) (@vinhnx)
#### Other
- Update TODO (7d612523)
#### Refactors
- Improve input rendering style handling and add ANSI color support (73a715b7)
## 0.134.13 - 2026-07-06
### Highlights
### Other Changes
#### Other
- Update TODOs (9152b2d3)
## 0.134.12 - 2026-07-06
### Highlights
### Other Changes
#### Refactors
- Clean up unused compiler settings and warnings in build configurations (129517c9) (@vinhnx)
## 0.134.11 - 2026-07-06
### Highlights
#### Features
- Add MultiErrors type implementing error-parameter pattern (145ea479)
### Other Changes
#### CI
- Update workflows to use install-action for dependencies and enable sccache (0f3af6e4)
#### Other
- Update notes (68cf3fc3)
- Update TODO (8ae6078c)
#### Refactors
- Enhance testing commands and scripts for improved efficiency (c7fcffd1)
## 0.134.10 - 2026-07-06
### Highlights
### Other Changes
#### Other
- Update help cli (114dbecd)
- Refactor provider noise handling and harmony sanitization (4c402a60)
## 0.134.9 - 2026-07-05
### Highlights
#### Bug Fixes
- Resolve apply_patch format confusion and updater stdout leak (53502b15)
- Cross-map max_results/globs/format in unified_search action=list (a8513c01)
- Resolve unified_search action=outline parameter confusion and refactor (be92673d)
### Other Changes
#### Other
- Enhance ast-grep code outline (8d2278b8)
## 0.134.8 - 2026-07-05
### Highlights
#### Bug Fixes
- Remove incorrect cfg gate on env_lock import (breaks Linux/Windows CI) (bb0c7d1f) (@vinhnx)
### Other Changes
## 0.134.7 - 2026-07-05
### Highlights
#### Bug Fixes
- Retain loop-detection metadata in compacted tool output (9bfed42a)
### Other Changes
#### Other
- Refactor code to allow missing documentation warnings and improve error handling in tests and components (defef020)
- Remove unused model override warnings and related test (ab912466)
- Fix em-dash guidance string, improve user message retention in compaction, and update test assertions (9c7f804c)
## 0.134.6 - 2026-07-04
### Highlights
### Other Changes
#### Other
- Fix/tool budget (05b182c9) (@kernitus)
## 0.134.5 - 2026-07-04
### Highlights
#### Documentation
- Document user AGENTS loading (bb56bfa3) (@kernitus)
### Other Changes
#### Other
- Add tracking state and turn metrics for agent sessions; implement tool registry interfaces and trait implementations (c624c3a8)
- Enhance tool registration and UI components with detailed documentation and new features (44f400bc)
- Enhance tool call ID generation and validation across providers; update tests for uniqueness and consistency (0970ee36)
- Add parser infrastructure for textual tool calls and enhance argument parsing (2dc72e9b)
- Refactor read family cap logic to distinguish paginated reads and prevent false trip conditions (d7dc7404)
- Refactor text tool parsers to use ParseResult for improved error handling and add consumed span tracking (eca6ddd8)
- Implement interactive API key prompt during first-run setup and enhance error guidance for authentication issues (e039e878)
- Update TODO (2ffd4a00)
#### Refactors
- Streamline test setup for loop detection and turn metrics (c12ac368)
- Update unwrap/expect/panic handling to use context for better error messages (c7268b71)
- Simplify task and context conversion using iterator methods (9c162eca)
- Improve JSON parsing in tests for better error handling (0198b3ba)
- Implement actor pattern for async tasks and improve channel handling (83e78171)
- Optimize string formatting and memory allocation in various modules (b79564e4)
- Replace string formatting with writeln! for improved performance and readability (83bc46d4)
- Streamline tool name normalization and enhance test coverage for apply_patch and file operations (998e762c)
## 0.134.4 - 2026-07-04
### Highlights
#### Documentation
- Update per-crate AGENTS.md for DRY refactor conventions (daaec30c)
#### Features
- Enhance error handling with failure classification and guidance (93d5565c)
- Implement JSON schema support for string-backed identifier newtypes (62fc40c4)
### Other Changes
#### Other
- Fix/keyboard input & image pasting (be4b05b3) (@kernitus)
- Add tool definition token tracking and improve tool search indexing (65a12f25)
#### Refactors
- Unify workspace path-containment checks into vtcode-commons (48e7d3f6)
- Consolidate error classification onto ErrorCategory (f3372ba6)
- Consolidate retry policy math into vtcode-commons (e0e49b84)
- Table-drive ModelId string/provider/parse matches (d1acff00)
- Introduce OpenAiCompatSpec shell and migrate Moonshot provider (1a877fdd)
- Migrate Poolside provider onto OpenAiCompatSpec (60937ec1)
- Migrate Mistral provider onto OpenAiCompatSpec (4affecb2)
- Migrate Qwen provider onto OpenAiCompatSpec (43c2bd17)
- Migrate MiMo provider onto OpenAiCompatSpec (fdf95528)
- Migrate StepFun provider onto OpenAiCompatSpec (bb897ff7)
- Migrate Z.AI provider onto OpenAiCompatSpec (5de6f37e)
- Migrate DeepSeek provider onto OpenAiCompatSpec (2d20a102)
- Fold OpenCode providers onto OpenAiCompatCore (c39d3d08)
- Migrate Evolink provider onto OpenAiCompatSpec (409e6a8f)
- Introduce id_newtype for plugin and marketplace identifiers (fe557352)
## 0.134.3 - 2026-07-01
### Highlights
#### Bug Fixes
- Revert agent-client-protocol 0.10.4 and rusqlite 0.35 to fix 0.134.2 build (f8dddd4f)
#### Features
- Migrate agent-client-protocol from 0.10.4 to 1.0.1 (SACP builder API) (210b1ca5)
### Other Changes
## 0.134.2 - 2026-07-01
### Highlights
#### Bug Fixes
- Downgrade notify-rust to 4.17.0 for rustc 1.88.0 compat (797fa278)
- Adapt to rmcp 2.0.0 breaking API changes (c4501adf)
### Other Changes
## 0.134.1 - 2026-07-01
### Highlights
#### Bug Fixes
- Scope recovery cap to current turn (66216d16) (@kernitus)
### Other Changes
## 0.134.0 - 2026-07-01
### Highlights
#### Bug Fixes
- Downgrade subagents visibility from pub(crate) to pub(super) (6c1da4d4)
- Add xhigh/max effort picker support for Sonnet 5, Fable 5, Mythos 5; fix adaptive thinking profiles per Anthropic docs (9eff2382)
#### Documentation
- Refresh stale structural_search/subagents file paths after split (4a1d7283)
#### Features
- Add Claude Sonnet 5, Fable 5, and Mythos 5 to Anthropic provider (12ebfcc7)
### Other Changes
#### Other
- Revert "refactor: remove Claude Fable 5 references and update to Claude Opus 4.8 across models and capabilities" (537a0e75)
#### Refactors
- Split structural_search.rs into focused submodules (717f75f7)
- Split subagents/mod.rs impl into controller submodules (f5203b6e)
## 0.133.23 - 2026-06-29
### Highlights
### Other Changes
#### Other
- Refactor code structure for improved readability and maintainability (643f7d2c)
## 0.133.22 - 2026-06-28
### Highlights
#### Bug Fixes
- Stream isolated review subcalls (a7ffd7f7) (@kernitus)
- Ignore responses keepalive events (f33650f5) (@kernitus)
- Update compatible responses retry test (7b73ba2c) (@kernitus)
#### Documentation
- Add web_search tool reference, document /continue and did-you-mean (90a321b2)
- Expand TODO with known tech debt items (144491a0)
- Add defuddle_fetch tool reference and web_search config fields (5040faca)
- Fix stale src/tui.rs references, update CONTRIBUTING.md and INDEX.md (8b0518e1)
- Audit and trim per-crate AGENTS.md files (018656ea)
- Align CI/CD docs with actual workflow files (5e39c041)
#### Features
- Implement cross-turn tracking and limit assistant text responses to prevent runaway loops (466c0c24)
- Add tiktoken integration for token counting and hierarchical summarization (3d5af4ed)
### Other Changes
#### Other
- Convert stringly-typed ResponseId, OutputItemId, ContentPartId to newtypes (2df60016)
- Implement recovery fallback messages and enhance outline search with directory summaries (75429ee7)
- Implement few-shot example selection and rendering; enhance tool descriptions and documentation. (82f078bb)
- Add persistent audit logging and per-MCP-server sandboxing (086208e3)
- Implement async approval blocker and escalation gate for irreversible actions (32d15be6)
- Add task history, message metadata, pending actions, and state schema modules (b670e3a0)
- Enhance tool descriptions for clarity and guidance on usage (b071d6a5)
- Enhance agent execution flow with escalation handling and stagnation detection (209364a4)
- Optimize token count management and enhance loop detection efficiency (e1f90cec)
- Refactor escalation decision enum and clean up evaluator gate outcome handling (c27527db)
- Refine system prompt guidelines for clarity and simplicity in agent operations (ae9a8a81)
- Clarify description for 'follow' parameter in unified search schema (e86300cc)
- Enhance skill availability messages to include instructions for listing full catalog (5512d9df)
- Refactor tool outcome handling and improve skill overflow messages for clarity (782b8fc7)
- Refactor model ID matching and clean up import statements for clarity (764128be)
- Refactor string formatting for improved readability and consistency across tests and core modules (87a375f3)
#### Tests
- Fix prompt cache retention fixtures (6d032016) (@kernitus)
## 0.133.21 - 2026-06-28
### Highlights
### Other Changes
#### Other
- Update run (a2f82005)
## 0.133.20 - 2026-06-28
### Highlights
#### Bug Fixes
- Deserialize StatusLineMode case-insensitively (2badec48) (@RobertBorg)
- Prevent loop detection from returning empty results and strengthen blocking (db5d7d5d)
#### Features
- Enhance Miri checks and add safety limits for Copilot payloads (352d610f)
- Add web_search, defuddle_fetch, and TOML-driven network allowlist (1fb4b50d)
- Add outline mode for unified_search to generate symbol maps (c12d2c53)
- Update TUI fixes, ast-grep auto-install, and review fixes (f0f0c066)
- Add binary override checks for ast-grep installation and testing (f5d2f650)
- Implement post-tool recovery cycle tracking and handling to prevent infinite loops (745e9488)
- Implement post-tool recovery cycle cap check and cleanup temporary files (1e36699e)
### Other Changes
#### Other
- Update README (fa0180f3) (@vinhnx)
- Enhance compatibility and error handling across various modules (d5249d00)
#### Refactors
- Reduce default output limits and enhance loop detection for command tools (47eab8f4)
## 0.133.19 - 2026-06-27
### Highlights
### Other Changes
#### Other
- Update TODO (bde8fd7e)
- Revert "chore: enhance build scripts for parallel processing and artifact management" (d6421236)
- Revert "Fix release script" (6ec480e3)
- Add new models and update configurations for OpenCode Go (b8e3282e)
## 0.133.18 - 2026-06-27
### Highlights
#### Bug Fixes
- Pin rmcp to =1.7.0 to prevent Arc<InitializeResult> type mismatch (e25976ae)
#### Features
- Colorize 'did you mean?' suggestions (15f5e609)
### Other Changes
## 0.133.17 - 2026-06-26
### Highlights
### Other Changes
## 0.133.16 - 2026-06-26
### Highlights
#### Features
- Add 'did you mean?' suggestions for unrecognized commands (46e6b56b)
- Add /continue command to resume the most recent session (67fb16d5)
### Other Changes
#### Other
- Update README to streamline provider information (87b53bb7) (@vinhnx)
- Implement Git worktree management and loop memory persistence (e2fec1b7)
- Update documentation to include CLAUDE.md references alongside AGENTS.md (c76b9c93)
- Update TODO (30307681)
## 0.133.15 - 2026-06-26
### Highlights
*No highlighted changes*
### Other Changes
#### Chores
- enhance build scripts for parallel processing and artifact management (452221875) (@vinhnx)
#### Other
- Fix release script (947307b26) (@vinhnx)
### Contributors
@vinhnx
## 0.133.13 - 2026-06-26
### Highlights
*No highlighted changes*
### Other Changes
#### Refactors
- move inject_latest_memory_envelope to test configuration (991fc083f) (@vinhnx)
#### Other
- Update TODO (87a03136f) (@vinhnx)
- Update TODO (873af9c63) (@vinhnx)
### Contributors
@vinhnx
## 0.133.12 - 2026-06-25
### Highlights
### Other Changes
## 0.133.11 - 2026-06-25
### Highlights
#### Bug Fixes
- Shell policy byte-offset panic, glob compilation, and structural fallback bugs (a6c8ce38) (@vinhnx)
- Security hardening, concurrency fixes, and async I/O improvements (e0fcd0fa) (@vinhnx)
- Stop memory envelope refresh mutating live history (172fc2ed) (@kernitus)
- Make memory envelope refresh history-immutable (6f33e7dc) (@kernitus)
- Preserve full resume history verbatim (5d967855) (@kernitus)
- Resume progress archives from full history (b506f33a) (@kernitus)
- Refine validation for format, report_style, and builtin_rules in scan workflow (2edbfd55) (@vinhnx)
### Other Changes
#### Other
- Update gitignore (ab279a6d) (@vinhnx)
## 0.133.10 - 2026-06-24
### Highlights
#### Bug Fixes
- Prevent false loop detection when read extent differs (26113d9c) (@vinhnx)
- Add read-extent check to find_duplicate_in_history (fde336dc) (@vinhnx)
- Prevent false loop detection when read extent differs (823825a6) (@vinhnx)
- Surface edit-size limits, fix error messages, and improve task tracker continuity (80d1585a)
#### Documentation
- Track Rig upstream PRs and update stale version references (1431c9aa)
### Other Changes
#### Other
- Update TODO (b31383a4) (@vinhnx)
## 0.133.9 - 2026-06-24
### Highlights
#### Bug Fixes
- Preserve responses stream call ids (bda2d6fb) (@kernitus)
- Preserve responses stream call ids (a87f5006) (@kernitus)
- Classify responses stream rig gaps (c769df4b) (@kernitus)
- Preserve responses rig gap payloads (31d60dcb) (@kernitus)
- Preserve custom tool input stream payloads (e8175f2a) (@kernitus)
- Align responses stream parser policy (ec737084) (@kernitus)
- Avoid duplicate tool call replays (e3eaf4e1) (@kernitus)
- Preserve rig-unknown responses items (b5345844) (@kernitus)
- Preserve responses stream mode (9c301493) (@kernitus)
### Other Changes
#### Build
- Test rig 0.39 compatibility (0519913d) (@kernitus)
#### Other
- Update TODO (74789d35)
- Update TODO (ea9d8fe8)
#### Refactors
- Add responses adapter boundary (9486046c) (@kernitus)
- Route chatgpt requests through rig (2c0c5223) (@kernitus)
- Adapt openai responses sse via rig (5456eebc) (@kernitus)
- Register tools through rig toolset (a0c04d08) (@kernitus)
- Consolidate openai chatgpt responses glue (45077f86) (@kernitus)
- Document chatgpt rig request boundary (38198cde) (@kernitus)
#### Tests
- Cover responses status marker events (b8797d82) (@kernitus)
- Preserve responses stream conversions (de597a96) (@kernitus)
- Use documented code interpreter stream event (53c3559d) (@kernitus)
## 0.133.8 - 2026-06-24
### Highlights
#### Bug Fixes
- Improve ast-grep structural search reliability and YAML safety (5eb728bb)
- Prevent loop detection from killing entire turns via blocked-streak fuse (b773e66c)
### Other Changes
#### Other
- Improve search (882b6c2b)
- Improve unified_search family key differentiation and deny in-place editing commands (9d449d37)
- Add up/down arrow archived history navigation and structural search exclude parameter (7cff539f)
## 0.133.7 - 2026-06-24
### Highlights
#### Bug Fixes
- Allow selecting auto agent from mode palette (2efcf777)
### Other Changes
#### Other
- Refine (6948d078)
## 0.133.6 - 2026-06-23
### Highlights
#### Bug Fixes
- Address turn_525 planning-mode loop issues (3e165b98)
### Other Changes
#### Other
- Exclude TODO (6c3c1fb9) (@vinhnx)
- Update docs on provider config schema (a65c6e56) (@vinhnx)
- Exclude TODO from docs map (8069bf01) (@vinhnx)
- Add Z.ai GLM Coding Plan section to README (5d34b122) (@vinhnx)
- Enhance pattern detection with explicit ToolCallRecord struct and update feature vector documentation (12b26ed5)
- Refactor web fetch tool description and add tests for description validation (9958eb2e)
- Add strsim dependency and improve fuzzy matching in tool discovery (6f93ba51)
- Add structured error handling and content truncation for web fetch tool (ccb2a1cc)
- Implement cleanup functions for old temp spools and audit logs to manage disk usage (d4486c25)
- Refactor concurrency limits and cleanup functions for improved log management (3cc14f3b)
- Implement age-based cleanup for temp files in web fetch tool and update tests (786fc39d)
## 0.133.5 - 2026-06-22
### Highlights
#### Bug Fixes
- Handle TUI mode in prompt capability check (7582512e)
- Prevent TUI corruption when cycling agents via Tab (704b65c7)
- Guard raw println/print calls that corrupt TUI display (61baed0d)
- Drain terminal events to prevent input corruption after external app execution (a461c329)
#### Documentation
- Update TUI println audit with fix status and remaining items (2b039c28)
#### Features
- Implement domain-scoped approval caching for web_fetch tool (70e5c760)
- Implement workspace configuration with root layer control (8e617121)
### Other Changes
#### Other
- Register web_fetch tool and implement HITL approval flow (36aed2e5)
- Refactor model display names to use Cow for dynamic strings; update provider handling and validation logic (7b72264b)
## 0.133.4 - 2026-06-21
### Highlights
### Other Changes
#### Refactors
- Refactor: streamline theme resolution and update preference logic (1d1ecc31)
## 0.133.3 - 2026-06-21
### Highlights
#### Bug Fixes
- Preflight respects release channel, DRY config check, collapsible if (db12a590)
- Resolve clippy warnings across workspace (967c2e7d)
- Explicit semver equality guard prevents spurious update notices (e0573aa6)
#### Features
- Self-update check on every launch (4787b4d9)
### Other Changes
#### Other
- Simplify preflight debug logging, remove redundant variable (c2bb6672)
#### Refactors
- Update ATIF metrics handling and exit data structure for zero-allocation (908b5224)
## 0.133.2 - 2026-06-21
### Highlights
#### Bug Fixes
- Improve concurrency safety and fix race conditions (e270c3d6)
- Additional concurrency and safety improvements (6865b65d)
- Error handling, logic bugs, and integer overflow prevention (ad181e85)
- Replace expect() with poison recovery in search_runtime and cached_executor (14dc8a55)
- Wait 10ms for in-flight terminal responses in drain (defa765a)
- Add dead code annotations for unused methods and fields (4c200cd1)
### Other Changes
## 0.133.1 - 2026-06-21
### Highlights
#### Bug Fixes
- Drain terminal responses before disabling raw mode on exit (0b3a6e02)
- Resolve clippy issues and improve shell approval patterns (d78c2796)
- Improve raw display and messaging for budget exhaustion in recovery decision (5a844757)
- Resolve clippy issues across vtcode-exec-events, vtcode-mcp, vtcode-ui, vtcode-config (637942df)
- Clean up recovery messaging and improve tool call handling (3079b762)
- Unify recovery cleanup info message wording (9975023b)
- Remediate modes/subagents audit and clear pre-existing test drift (6620f73c)
#### Documentation
- Add model list to vtcode.toml.example (c8538c10)
#### Features
- Enhance tool policy and session management with interrupt handling and approval caching (bdb857b5)
- Enhance tool call guards and approval handling with new patterns and configurations (0ece95d4)
### Other Changes
#### Other
- Refactor numeric calculations to prevent sign loss and improve precision in various modules (9c5c7bc9)
## 0.133.0 - 2026-06-20
### Highlights
#### Bug Fixes
- Reduce mid-run stops and eliminate wasted duplicate tool calls (884b39b0)
- Reduce mid-run stops and eliminate wasted duplicate tool calls (a925cb80)
- Silently ignore floating overlay Selection events when no palette is active (e67db4e4)
- Stop constant writes of background_subagents.json (5cfc6c10)
#### Features
- Add Z.ai GLM Coding Plan with subscription details to README (593ba1ba) (@vinhnx)
- Enhance verification handling and refactor tool execution methods (d4f91648)
- Implement duplicate tool call detection and enhance recovery handling (89e92b27)
- Enhance agent selection and permissions handling for planning workflow (5c372a64)
- Enhance planning workflow with real-time header updates and tool output handling (16c55cfc)
- Improve planning confirmation flow and enhance tool output handling (49d64115)
- Enhance planning workflow with trust checks and tool behavior adjustments (be0b38e9)
- Implement tool policy overrides and enhance denial diagnostics in planning workflow (0cf67cf7)
- Merge session and subagent lifecycle events in primary hook configuration (ba264332)
- Prevent TUI interrupt from exiting the program and improve signal handling (8cac5ee2)
- Enhance planning workflow with agent type handling and patch file extraction (25227d31)
- Deactivate planning workflow on agent switch and refine exit triggers (4e95879b)
- Add mode command to switch active agent mode and implement mode palette (ef1b52b6)
- Enhance agent mode handling and improve tool output formatting (02c5a155)
- Add kimi-k2.7-code, glm-5.2, minimax-m3 models (ce6c7222)
- Improve signal handler responsiveness and refine tool action filtering (36da47e7)
- Enhance Ctrl+C handling and add agent context to tool logging (474fcdfb)
- Implement tool-free recovery mode switch and refactor related logic (a379f558)
- Enhance workspace detection and logging for safe-mode prompts (ed91aee6)
- Add support for native inline compaction in LLMProvider and related providers (f8146fce)
### Other Changes
#### Other
- Remove references to Nemotron 3 Super model from Ollama provider and related tests (e8a7077c) (@vinhnx)
- Refactor documentation links and update references across multiple files (8838ff0e) (@vinhnx)
- Update documentation for VT Code version 0.132.0, including new modules, features, and provider integrations. (1a2fe628) (@vinhnx)
- Revise VT Code description for clarity and conciseness (7a3b9abb) (@vinhnx)
- Rename sections for OpenAI and Anthropic in guide (f801db7d) (@vinhnx)
- Add global read-only budget limit and update harness limits prompt (7f96113d)
- Implement subagent mode with tighter read-only budgets and navigation limits (e7c2f746)
- Update TODO (69b98ae2)
- Update notes (8c0d97a8)
- Implement adaptive budget recovery mechanism and related tool definitions (1f518af1)
- Refactor manual compaction support and improve provider interface (29030ca3)
#### Refactors
- Update tool references from `grep_file` and `read_file` to `unified_search` and `unified_file` across documentation (87391096) (@vinhnx)
- Clean up code formatting and improve readability in multiple files (f5b06c97)
- Streamline fallback handling and improve code readability in tool execution (a6b542f4)
- Consolidate HTTP client and optimized client code, enhance command argument handling, and improve read-only command detection (6d3915db)
- Consolidate duplicated tool dispatch patterns and add fallback validation (0b37a840)
## 0.132.0 - 2026-06-16
### Highlights
#### Bug Fixes
- Correct logo path and improve formatting in README (2c6952e5)
#### Documentation
- Add signal handling architecture documentation (3d30aa17)
- Update tree detail rendering to use correct prefix styling (5f335b9d)
- Update tree detail rendering to use correct prefix character (50bb4f4a)
- Update safety documentation link to SECURITY_MODEL.md (5919991d)
- Update logo and fix documentation links in README (47c2d476)
#### Features
- Add include_cache_metrics parameter to stream handling functions and update related tests (200090ba)
- Add GLM-5.2 model support and update related configurations (a7ab54fb)
- Add GLM-5.2 model support and update related configurations (603720b2)
### Other Changes
#### Other
- Update README.md (f04686e6) (@vinhnx)
- Update README.md (a391ca62) (@vinhnx)
- Update TODO.md (d7aba241) (@vinhnx)
- Delete crate-consolidation-plan.md (60a4f9d2) (@vinhnx)
- Refactor chat completion URL handling and improve error formatting across providers (44a1f2ff)
#### Refactors
- Simplify stream handling in DeepSeek, StepFun, and ZAI providers (365e4a6a)
- Update reasoning_field to reasoning_fields for improved flexibility in stream handling (3fe34776)
## 0.131.1 - 2026-06-14
### Highlights
#### Bug Fixes
- Add libc dependency and create no-console-log test and snapshot files (a6c4b966)
- Update page size to read all 7 bytes in detailed file handling (1fa15165)
- Remove redundant 'action' parameter from noise filter (ff93faf2)
#### Features
- Enhance file reading guidance and tracking for unified_exec commands (82628462)
- Implement byte-range reading for large files with offset and page size parameters (46faf766)
### Other Changes
#### Other
- Update TODO (9357e9e6)
#### Refactors
- Streamline file reading logic and enhance line-numbered output (026b760d)
- Replace truncate_middle with truncate_path_middle for better path handling (42e0b8b1)
## 0.131.0 - 2026-06-14
### Highlights
#### Bug Fixes
- Remove non-publishable crates from publish list and fix cargo doc --all-features flag (41b65ad0)
### Other Changes
## 0.130.0 - 2026-06-14
### Highlights
### Other Changes
#### Other
- Refactor crate lists in publish and release scripts for clarity and consistency (57fbe342)
## 0.129.0 - 2026-06-14
### Highlights
#### Bug Fixes
- Update vtcode-core dependency to use version for publishing (c11ce2aa) (@vinhnx)
- Correct cargo test instructions and add x-ai/grok-4-fast:free to openrouter model support (0432d820) (@vinhnx)
- Stabilize iocraft command handling (396c2528) (@vinhnx)
- Specify version requirement for vtcode-core dependency to fix crates.io publishing (b298ac54) (@vinhnx)
- Restore transcript scroll and clean placeholder (f66a744a) (@vinhnx)
- Restore slash help and respect manual scroll (486d312c) (@vinhnx)
- Restore mouse capture guard (eaee5593) (@vinhnx)
- Update status bar help message for clarity (686327b9) (@vinhnx)
- Update provider and model settings to use OpenAI's gpt-5-codex (a02fce8c) (@vinhnx)
- Update provider and model settings in configuration (26ca4180) (@vinhnx)
- Update provider and model settings, add VHS demo guide, and replace demo image (0fe47f11) (@vinhnx)
- Update provider and model settings, add VHS demo guide, and replace demo image (6e3e7ca0) (@vinhnx)
- Correct codex tool calls and mouse selection (aa329625) (@vinhnx)
- Include response tool names for openai (bad9d93e) (@vinhnx)
- Include tool names in OpenAI responses payload (2e22ae5d) (@vinhnx)
- Update provider and model settings in configuration (21041709) (@vinhnx)
- Update provider and model settings, add VHS demo guide, and replace demo image (b76ffd86) (@vinhnx)
- Update OpenAI provider settings and adjust session bootstrap structure (cdfe4d52) (@vinhnx)
- Correct release asset names for Homebrew compatibility (c7e503a5) (@vinhnx)
- Improve MCP client initialization error handling (127da7dd) (@vinhnx)
- Remove undefined function call in release script (cea8b973) (@vinhnx)
- Use platform-specific process_group API for proper cross-platform support (18ad95c5) (@morler)
- Update model references to remove version suffix for Claude Sonnet 4.5 (0a8e01f5) (@vinhnx)
- Fix tests, deepseek integration (96febe2d) (@oiwn)
- Remove invalid workspace field from release.toml (81f9a888) (@vinhnx)
- Remove unsupported fields from release.toml (ca2e30b5) (@vinhnx)
- Unify auto flag handling (7d0b64d8) (@vinhnx)
- Update sed commands in Homebrew formula to use alternate delimiter (e576a780) (@vinhnx)
- Update sed commands in Homebrew formula to use alternate delimiter (4a975c98) (@vinhnx)
- Respect custom provider api keys (efa4bf75) (@vinhnx)
- Style section headers (6f9f29fc) (@vinhnx)
- Resolve duplicate manifest entry and iterator type (7532698c) (@vinhnx)
- Unify onboarding sections (0d2d4bb0) (@vinhnx)
- Clarify onboarding headings and code styling (b7014e38) (@vinhnx)
- Inject context7 api key configuration (bcb254f3) (@vinhnx)
- Remove context7 api key requirement (084f0389) (@vinhnx)
- Surface welcome highlights in header (ade64062) (@vinhnx)
- Update deps and clean justify lint (fba00928) (@vinhnx)
- Simplify welcome header meta (126cafef) (@vinhnx)
- Update GitHub Packages authentication checks and improve .npmrc configuration (34d0115f) (@vinhnx)
- Rename grep_file to grep_search in tool policy (8a34ffa4) (@vinhnx)
- Resolve duplicate reasoning display and ensure assistant message visibility (424820e4) (@vinhnx)
- Update agent configuration to use OpenAI provider and models (6eb024a5) (@vinhnx)
- Update agent configuration to use OpenRouter provider and models (387f25c4) (@vinhnx)
- Update agent configuration to use OpenAI provider and models (d1ac24f5) (@vinhnx)
- Downgrade openai-harmony dependency version to 0.0.3 (65854a98) (@vinhnguyenxuan-ct)
- Update modal list highlight symbol for improved visibility (dedb2f2e) (@vinhnx)
- Treat allowed directories as recursive (71795ebf) (@vinhnx)
- Refine tool approval selection logic and improve prompt display (6bc4e24a) (@vinhnx)
- Skip rendering internal fields in JSON output and improve debug logging (9e7268f1) (@vinhnx)
- Remove suppressed error logging for provider errors in agent loop (8736e504) (@vinhnx)
- Correct border rendering for curl command output (468e80e2) (@vinhnx)
- Remove mdbook workflow causing CI failure (3c3f6018) (@vinhnx)
- Add missing API key header in generate_stream method (6bbc6189) (@vinhnx)
- Add permissions section to workflow files (873a252b) (@vinhnx)
- Remove unnecessary newline in CI workflow (be34d75d) (@vinhnx)
- Correct base URL in MinimaxProvider configuration (e8844996) (@vinhnx)
- Update LLM provider and related settings to use OpenRouter (f8b97168) (@vinhnx)
- Adjust max_tool_loops to 20 and correct prompt file paths (78662775) (@vinhnx)
- Add Debug trait to InlineTextStyle for improved logging (fc6ce931) (@vinhnx)
- Correct tool name from run_command to run_terminal_cmd (ae2a0e44) (@vinhnx)
- Update tool name in test and improve conversation compression logic (bc8520d6) (@vinhnx)
- Update language model tool properties for VTCode Update Plan (342d6bd6) (@vinhnx)
- Rename VTCode Update Plan tool for consistency (f573248f) (@vinhnx)
- Update changelog generation to handle date formatting correctly (9e2303eb) (@vinhnx)
- Restore npm/package.json file removed in error (6332545f) (@vinhnx)
- Add wasm32-wasip2 component to toolchain configuration (1bbcb0e7) (@vinhnx)
- Update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies (a0cb4a0d) (@vinhnx)
- Remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml (9732a469) (@vinhnx)
- Ensure publish is set to false in Cargo.toml (5938a4f0) (@vinhnx)
- Update mcp-types version to 0.1.1 in Cargo.toml (925054bc) (@vinhnx)
- Update mcp-types version to 0.1.1 in Cargo.lock (009685b0) (@vinhnx)
- Ensure publish is set to false for mcp-types in release.toml (64ec55e6) (@vinhnx)
- Remove mcp-types configuration from release.toml (020117c6) (@vinhnx)
- Remove jsonschema dependency from mcp-types in Cargo.lock (8a80a605) (@vinhnx)
- Add Debug trait to MessageStyle enum (0b7f1cda) (@vinhnx)
- Update Moonshot model references from KIMI_K2_THINKING_HEAVY to KIMI_K2_THINKING_TURBO (aead34f1) (@vinhnx)
- Escape newlines properly in sed command for macOS (ac146436) (@vinhnx)
- Use awk instead of sed for changelog updates on macOS (19fbaabc) (@vinhnx)
- Use perl instead of awk for changelog updates on macOS (7869f422) (@vinhnx)
- Use temporary file approach with awk for changelog updates on macOS (3446acdc) (@vinhnx)
- Redirect print functions to stderr to avoid command substitution issues (c0d60187) (@vinhnx)
- Revert version in package.json to 0.42.13 (cc26fb26) (@vinhnx)
- Update tool policies and disable time provider in configuration (fe10c9b5) (@vinhnx)
- Remove unused import in code_executor (9333d865) (@vinhnx)
- Resolve compilation warnings and duplicate test module (11f76b3d) (@vinhnx)
- Revert extension.toml to valid Zed format (4cb061ac) (@vinhnx)
- Update LLM provider and models to use Ollama (cde009cc) (@vinhnx)
- Optimize list_files tool for improved pagination and reduce default page size (62b918d9) (@vinhnx)
- Redirect logging to stderr to prevent stdout pollution in install script (530c8550) (@vinhnx)
- Update Claude model identifiers and descriptions for accuracy (ebf10f14) (@vinhnx)
- Update lmstudio provider: remove stale 'For now' comment and simplify validation; update related utility and policy files (d033f08b) (@vinhnx)
- Truncate verbose reasoning output to reduce noise during tool execution (a248b138) (@vinhnx)
- Eliminate duplicate response output caused by self-review logic (655c663b) (@vinhnx)
- Improve tool failure handling by tracking failed attempts (ee7d0dde) (@vinhnx)
- Resolve clippy warnings (range_contains, doc comments, identical blocks) (5f7a58da) (@vinhnx)
- Prevent infinite tool loops by using >= instead of > (ca5a66a5) (@vinhnx)
- Suppress dead_code warnings for intentionally disabled features (27818264) (@vinhnx)
- Remove duplicate user message in turn loop (8f28c05d) (@vinhnx)
- Remove duplicate user message from conversation history (7c7a216b) (@vinhnx)
- Sync embedded asset for generate-agent-file.md (ab337571) (@vinhnx)
- Improve loop detection for repeated tool calls (0999db9b) (@vinhnx)
- Remove overly complex sandbox cache clearing on PTY retry (0f415ee4) (@vinhnx)
- Ensure development tools are always in PATH with fallback paths (4d81f564) (@vinhnx)
- Expose shell tool to LLM by setting expose_in_llm to true (464265fe) (@vinhnx)
- Remove duplicate test block with non-existent method in zed.rs (e3501a92) (@vinhnx)
- Add environment() and path() to EditorBuilder to properly detect and launch editor (21ecb580) (@vinhnx)
- Apply Ratatui FAQ best practices - fix async/tokio issues (5fe91969) (@vinhnx)
- Track fire-and-forget tokio::spawn tasks with JoinHandles (545d1817) (@vinhnx)
- Improve `edit_file` tool's robustness (028fdb36) (@vinhnx)
- Update public re-export and documentation to reference file_helpers instead of legacy (b1b072f4) (@vinhnx)
- Clear thinking spinner message on all agent response command types (a5a0a797) (@vinhnx)
- Move thinking spinner display to after user message in transcript (6df6d5cc) (@vinhnx)
- Revise thinking spinner message to use first-person agent voice (5a01e79e) (@vinhnx)
- Resolve all cargo clippy warnings and update rust toolchain to stable (341cd5c6) (@vinhnx)
- Correct RUSTFLAGS invalid option and align system prompt with actual tool definitions (d7060e15) (@vinhnx)
- Remove unused spawn_session import (aece5068) (@vinhnx)
- Gracefully handle provider API errors without panicking (865745f8) (@vinhnx)
- Clear spinner before displaying error message (fd74210d) (@vinhnx)
- Improve error handling in LLM client creation and tool execution (49c256b8) (@vinhnx)
- Vtcode agent skill discovery using SkillLoader instead of SkillManager (f5f413c5) (@vinhnx)
- Add missing skill tool function declaration (390245e1) (@vinhnx)
- Improve skill tool output to include full instructions (171a43e7) (@vinhnx)
- Enhance skill discovery and loading functionality for vtcode agent (1b19b4cd) (@vinhnx)
- Remove outdated skill discovery documentation and integrate new skill loading functionality (167ff554) (@vinhnx)
- Update execute_code and skill policies to allow execution (a6eb17b9) (@vinhnx)
- Include templates directory in package for crates.io publishing (72cb14f6) (@vinhnx)
- Correct logical operator for XAI provider model check (b414681c) (@vinhnx)
- Handle errors in AtomicWriter creation and improve diff operations tests (a32348aa) (@vinhnx)
- Correct test expectations for token threshold boundaries (3a910682) (@vinhnx)
- Prevent arrow keys from triggering external editor launch (e6ebffa0) (@vinhnx)
- Remove external editor keybinding (Control+E) (16bf4737) (@vinhnx)
- Update tool policies and configuration settings for improved performance (9f16528e) (@vinhnx)
- Resolve GitHub release binary upload failures and enhance release script verification with new documentation. (6e350f53) (@vinhnx)
- Align OpenAI Responses API implementation with official spec (d8764701) (@vinhnx)
- Reorder MCP time policies and update tool policy documentation (3bddffbb) (@vinhnx)
- Change terminal output from stderr to stdout for ModernTui (d74fbcbf) (@vinhnx)
- Redirect terminal commands from stdout to stderr for TUI functionality (a224d80b) (@vinhnx)
- Fix: add missing package-lock.json (9fbcd6af) (@vinhnx)
- Unignore .github directory to enable GitHub Actions CI/CD workflows (6a0d9780) (@vinhnx)
- Update release workflow to handle npm publishing correctly (3a54970a) (@vinhnx)
- Improve npm publishing with token-based auth fallback (b808d8e3) (@vinhnx)
- Configure GitHub Packages authentication correctly (e2cbb95e) (@vinhnx)
- Rename npm package from vtcode-bin to vtcode (f54dd8b4) (@vinhnx)
- Remove npm installation due to GitHub Actions costs (1b6277a2) (@vinhnx)
- Disable npm publishing in release.sh (4f892d59) (@vinhnx)
- Disable JSON object output and Responses API for GLM models and refine streaming completion event content handling. (3e5520a8) (@vinhnx)
- Fix(models): Correct ClaudeOpus41 version and update related assertions (9e04c75c) (@vinhnx)
- Update dependencies and improve error handling in terminal functions (39494eef) (@vinhnx)
- Homebrew release automation and YAML workflow indentation (3fa45f4b) (@vinhnx)
- Improve homebrew formula regex patterns for reliable checksum updates (1d3d0b0f) (@vinhnx)
- Resolve python string interpolation in homebrew formula updates (e36fca34) (@vinhnx)
- Improve release.toml commit message template for consistency (43b5f8e2) (@vinhnx)
- Correct Python variable substitution in Homebrew workflow (a12861b1) (@vinhnx)
- Redirect all logging to stderr in installer script (db1a1105) (@vinhnx)
- Ensure get_download_url outputs only URL to stdout (3e93bb6d) (@vinhnx)
- Conditionally import debug-only items to fix release build (0bb253a2) (@vinhnx)
- Install OpenSSL dependencies for Linux builds in CI (1b74d554) (@vinhnx)
- Suppress unused_imports warning in openai.rs for CI compatibility (7199f351) (@vinhnx)
- Build-release workflow now triggers on tag push events (28687630) (@vinhnx)
- Suppress dead_code warnings for planned/stub functions (105c9869) (@vinhnx)
- Resolve Windows build errors in vtcode-core (c5ab4d74) (@vinhnx)
- Remove dead code and fix compilation errors (143d605c) (@vinhnx)
- Prefix unused variable with underscore in config_watcher (c7347a47) (@vinhnx)
- Suppress dead code warnings for unused UI and agent functions (3e906d63) (@vinhnx)
- Prevent duplicate reasoning output during finalization (693d5417) (@vinhnx)
- Resolve ambiguous AsRef trait for Cow<'_, str> (2e5d3dfe) (@vinhnx)
- Resolve ambiguous AsRef trait for Cow in zed.rs (99c4537f) (@vinhnx)
- Remove unused methods is_planner_active and is_coder_active (43bf61bc) (@vinhnx)
- Enhance output spooling logic for PTY commands and handle double-serialized JSON (02903155) (@vinhnx)
- Update allowed tools list to include request_user_input (549ff583) (@vinhnx)
- Address unresolved imports and modules in tests across multiple components (15b52042) (@vinhnx)
- Remove `check_output.txt` to clean up outdated and obsolete error logs (1b268b71) (@vinhnx)
- Update GitHub release title format and improve changelog generation (97571aa2) (@vinhnx)
- Streamline output handling in ask command and improve code extraction logic (8de09e16) (@vinhnx)
- Improve release fetching with fallback for older versions, enhance platform-specific binary handling (5d1344a8) (@vinhnx)
- Streamline platform-specific binary builds, refactor `ask` command implementation, and enhance local release workflow (d3bca9c2) (@vinhnx)
- Add spinner for long-running tasks, improve cursor handling, and streamline release fetching logic (484bfef1) (@vinhnx)
- Refactor ask command output handling, enhance pipeline detection, and centralize code extraction logic (b2aee725) (@vinhnx)
- Enhance checksum verification logic across scripts, add fallback for individual sha256 files, and improve error handling (5abd7c60) (@vinhnx)
- Enforce tools_policy prompts, refactor workspace trust application, and enhance command safety checks (394aa59b) (@vinhnx)
- Adjust color mappings, enhance markdown rendering logic, and add tests for new edge cases in tool policies and UI interactions (8309ffc1) (@vinhnx)
- Adjust tool policies, refactor text styling logic, and optimize message rendering indentation (8123ac34) (@vinhnx)
- Improve shimmer animation handling, refactor spinner updates, and enhance status rendering logic (0a40ad46) (@vinhnx)
- Add support for VTCODE_CONFIG_PATH, enhance configuration loading logic, and remove unused `.aiignore` file (bf6547aa) (@vinhnx)
- Update default theme and reasoning effort, improve config overrides, adjust workspace trust mode, and enhance contribution docs (85c1ba7f) (@vinhnx)
- Add persistence for editing and autonomous mode settings, and align theme with active configuration (bfea2f52) (@vinhnx)
- Add persistence for editing and autonomous mode settings, and align theme with active configuration (ecacbc19) (@vinhnx)
- Update default theme, enable todo planning, refine tool output and display settings, and adjust tool policies for streamlined configuration (dcc7043e) (@vinhnx)
- Update default theme, enable todo planning, refine tool output and display settings, and adjust tool policies for streamlined configuration (3308c57e) (@vinhnx)
- Refine markdown styling logic for strong, heading, and inline code elements, and enhance theme-based accent application (04278e00) (@vinhnx)
- Enable `unified_file` tool in the sandbox and refactor diff preview styling to use a color palette. (15715b9e) (@vinhnx)
- Preserve tool_exists when MCP tool check returns false (999f4a85) (@vinhnx)
- Replace println with tracing debug for git repository check (c6c11053) (@vinhnx)
- Correct file path conversion in log_access method (ec7e3dc4) (@vinhnx)
- Skip gh auth checks in dry-run mode (8f4f8dae) (@vinhnx)
- Skip hanging GitHub CLI refresh in release script (5bf61747) (@vinhnx)
- Skip hanging GitHub CLI refresh in build script (c77bb4aa) (@vinhnx)
- Rename directory with colon to be Windows-compatible (f533addc) (@vinhnx)
- Resolve duplicate model entries and correct legacy model references (a712a191) (@vinhnx)
- Update spinner finish behavior for cancellation handling (73a7d72f) (@vinhnx)
- Handle empty changelog updates gracefully (2d9ccac2) (@vinhnx)
- Resolve tool call ID mapping issue and update JSON handling in messages (9af9d34b) (@vinhnx)
- Add exit_plan_mode tool to planner agent and update tests (63a6835e) (@vinhnx)
- Use cargo metadata for reliable crate version retrieval (5df80eaf) (@vinhnx)
- Update versioning format to remove 'v' prefix in tags and URLs (081b6f73) (@vinhnx)
- Resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
- Correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- Prevent footer panic when hint is absent, refactor path argument to `&Path`, and optimize sidebar string truncation. (59525d4b) (@vinhnx)
- Update default model in configuration to glm-5:cloud (1700a7e4) (@vinhnx)
- Correct changelog generation to use the previous release tag instead of a fixed version. (b0437d27) (@vinhnx)
- Update chat input placeholders for clarity and improved user guidance (284a2390) (@vinhnx)
- Update chat input placeholders for clarity and improved user guidance (2e178d32) (@vinhnx)
- Disable scroll indicator in status bar (c5d4373d) (@vinhnx)
- Improve output messages for run completion and error handling (833564dd) (@vinhnx)
- Strip x-extension fields from tool schemas and handle empty safety blocks (7ff0b66f)
- Update contributor handling in changelog template (f1ef0a2c) (@vinhnx)
- Use RngExt trait for random_range method in slug.rs (b9b03932) (@vinhnx)
- Trim whitespace from custom config and data directory environment variables (179f1ef9) (@vinhnx)
- Use correct repo name and handle tags without v prefix (6dfb0356) (@vinhnx)
- Remove unused OpenRouter model variants and fix panic_hook warnings (d4afaa47) (@vinhnx)
- Update Cargo.toml to exclude all docs except CONFIG_FIELD_REFERENCE.md (6903748b) (@vinhnx)
- Update terminal notification handling for ghostty to use bell-only (ca512670) (@vinhnx)
- Preserve v prefix in GitHub release URLs for install script (3f2088be) (@vinhnx)
- Use exact tag names from GitHub API in install script (b22bd9af) (@vinhnx)
- Add missing environment variables for Rust compilation in Dockerfile (4a51a588) (@vinhnx)
- Update success message and fallback logic for OpenAI API key (3895d00d) (@vinhnx)
- Enhance OpenAI OAuth handling and improve error reporting (9dd80633) (@vinhnx)
- Improve plan mode entry and lifecycle tracking (13638d96) (@vinhnx)
- Improve argument and stderr preview truncation for safety (3a09f7c1) (@vinhnx)
- Ensure all OpenAI sessions are cleared on logout and fallback to API key (3d6cb033) (@vinhnx)
- Auto-copy highlighted TUI selections (5a33748a) (@vinhnx)
- Copy selected input on command-c (05a4472f) (@vinhnx)
- Remove repeated unrelated worktree prompts (8f93b519) (@vinhnx)
- Remove repeated unrelated worktree prompts (1f4ea2bd) (@vinhnx)
- Handle already cleared oauth logout states (081e20a6) (@vinhnx)
- Remove idle input notification (367e91ba) (@vinhnx)
- Tighten prompt cache continuity and retention handling (7afbe9f0) (@vinhnx)
- Update sha256 checksum for macOS ARM architecture (39e5ace9) (@vinhnx)
- Update macOS ARM checksum and enhance Homebrew publishing script (58d1989a) (@vinhnx)
- Stabilize Zed ACP startup and transcripts (72f91134) (@vinhnx)
- Preserve compaction context in prompts and summaries (b38f1877) (@vinhnx)
- Update external editor instructions to reflect single-click behavior (7f9f0557) (@vinhnx)
- Unify external URL guardrails across VT Code flows (a6444c4b) (@vinhnx)
- Suppress loading UI for spool chunk reads (1635a52f) (@vinhnx)
- Pass model behavior to anthropic api provider (70944e75) (@vinhnx)
- Correct post-release regressions (3cd9dde3) (@vinhnx)
- Encode SHA256 digest as hex manually (db7768f8) (@vinhnx)
- Encode config fingerprint digest as lowercase hex (4d470c5e) (@vinhnx)
- Update zig for ghostty release builds (b296ab3f) (@vinhnx)
- Tighten custom tool handling for OpenAI responses (3d5956b5) (@vinhnx)
- Preserve blocked hook prompt context (e07a9b50) (@vinhnx)
- Simplify ToolPolicy::Prompt handling in ToolRegistry (63f348e3) (@vinhnx)
- Standardize description quotes in workflow inputs (4ff9ec73) (@vinhnx)
- Add prctl feature to nix for Linux cross-compilation (c125d597) (@vinhnx)
- Streamline nix dependency features formatting (485f3b45) (@vinhnx)
- Include terminal-detection before tui publish (6f68a365) (@vinhnx)
- Publish specs crates before core (0c1da64d) (@vinhnx)
- Allow publishing specs crates (2f91bd14) (@vinhnx)
- Update task_tracker permission and model version in configuration (05405f5c) (@vinhnx)
- Harden perf baseline startup measurement (32f657dc) (@vinhnx)
- Animate active file operation indicator (c2baca0a) (@vinhnx)
- Enhance command safety checks and segment handling (d7491771) (@vinhnx)
- Dim top-level task tail lines in tool blocks (060ec1f5) (@vinhnx)
- Isolate shared state locks (0d685107) (@vinhnx)
- Clear publish warnings (e5d2cb78) (@vinhnx)
- Prevent blocked handoff on tool-free recovery tool calls (f3d5a7ee) (@vinhnx)
- Make external editor opens transient and suppress escape leaks (1e8bd327) (@vinhnx)
- Update theme configuration to use homebrew style (4205a879) (@vinhnx)
- Update agent configuration and tool policies for improved functionality (f9cdc03d) (@vinhnx)
- Refine asset selection logic to prioritize .tar.gz over .sha256 files (f113d8a6) (@vinhnx)
- Update condition for missing platform binaries in release script (3e88cd0e) (@vinhnx)
- Improve user prompt handling and update configuration settings (934ee662) (@vinhnx)
- Clear RUSTC_WRAPPER to avoid sccache permission issues during builds and cargo release (c39d09d5) (@vinhnx)
- Upgrade rustls-webpki to 0.103.12 (#2 #3) (b53f9c9a) (@vinhnx)
- Update default model to claude-sonnet-4-6 across configurations and documentation (13f942bb) (@vinhnx)
- Clear RUSTC_WRAPPER to prevent permission issues and ensure proper builds (eeb20ee6) (@vinhnx)
- Simplify metadata code retrieval in tests (5c4dbd5b) (@vinhnx)
- Handle dated GPT-5.5 alias in model routing (34079b94) (@vinhnx)
- Enable atlas cloud custom provider flows (84750086) (@lucaszhu-hue)
- Improve tool output rendering for command success with no output (de2f5bad) (@vinhnx)
- Ensure duration calculations do not exceed maximum limits (c0fecf23) (@vinhnx)
- Update stop hook to use final text for improved response handling (24739e82) (@vinhnx)
- Ensure blank lines are rendered as placeholders in diff output (2395d8ea) (@vinhnx)
- Record approval decisions for pattern learning in handle_builtin_permission (39cc8183) (@vinhnx)
- Remove invalid -C probe-stack codegen flag (74cd4eda) (@vinhnx)
- Update parameter types for setrlimit and getrlimit functions (700440e8) (@vinhnx)
- Remove setrlimit/getrlimit wrappers to fix type mismatch on Linux (eb2744d1) (@vinhnx)
- Update validation messages for structured outputs and task budget tokens in Claude Opus 4.8 (705f7def) (@vinhnx)
- Update API response thinking modes and configuration defaults for Anthropic provider (b814f5f5) (@vinhnx)
- Expose matches_model function for public use and update validation logic (4da619bf) (@vinhnx)
- Update model descriptions for Claude Haiku and Claude Sonnet to enhance clarity and detail (ee381fc1) (@vinhnx)
- Update provider checks to include additional model names (52a57555) (@vinhnx)
- Pin compatible protocol SDK (ec6399d2) (@vinhnx)
- Update API key and default model settings in vtcode.toml (fc4c1dd6) (@vinhnx)
- Cross-build platform compat, yanked core2 dep (a935bace) (@vinhnx)
- Build windows arm64 natively, cross images require ghcr auth (395c16d5) (@vinhnx)
- Run cross builds sequentially to avoid OOM on 16GB M4 (5ad249b4) (@vinhnx)
- Upload failure, homebrew update, skip aarch64 cross on mac (44ea5199) (@vinhnx)
- Skip Windows builds when targets missing, upload assets individually (ced93f5f)
- Resolve double escape functionality issue and improve response handling (da06eb28) (@vinhnx)
- Skip OS keyring during tests/CI to avoid Keychain prompts (a60f1721) (@vinhnx)
- Ensure deterministic hashing by sorting JSON object keys (38fe3afb) (@vinhnx)
- Add vtcode-ghostty-core to publish order before vtcode-core (eeb15246) (@vinhnx)
- Add vtcode-macros to publish order before vtcode-core (55f2e4df) (@vinhnx)
- Make crate publishing non-fatal in release.sh and add dependency validation to publish script (12ef4e7b) (@vinhnx)
- Remove double escape functionality and update related hints (4884c153) (@vinhnx)
- Suppress deprecation warnings for anstyle utilities (27aa88f9) (@vinhnx)
- Resolve 3 critical harness robustness bugs from session turn_364 (ebdbafb1) (@vinhnx)
- Resolve docs.rs build failure for vtcode-config (5f722204) (@vinhnx)
- Handle parameterised close tags in collect_pseudo_marker_regions (dac7afa0) (@vinhnx)
- Apply PR review fixes (d530f6c7) (@vinhnx)
- Restore harness robustness plan section in TODO.md (84a652ea) (@vinhnx)
- Update tool call region collection to use pseudo marker (c67d347c) (@vinhnx)
- Improve screenshot path resolver to stop consuming trailing text (2b101bb1) (@vinhnx)
- Improve screenshot path resolver and extract shared path utilities (75e445c4) (@vinhnx)
- Update path handling in trim_trailing_image_path_str function (59208038) (@vinhnx)
- Add image handling in request processing and refactor fallback logic (52f780ae) (@vinhnx)
- Pin ratatui to =0.30.0 to prevent CellEffect Send breakage (8539a774) (@vinhnx)
- Use token-based auth for homebrew-tap push instead of gh credential (66ac78b3) (@vinhnx)
- Pin ratatui-widgets to =0.3.0 to fix cargo install (c60e1663) (@vinhnx)
- Fallback to unauthenticated GitHub API on 401 (fdc666f9) (@vinhnx)
- Handle terminal clear failure during force redraw (b85bec6e) (@vinhnx)
- Update default primary agent from 'duck' to 'build' across configurations and tests (db0df331)
#### Documentation
- Refresh release status for v0.11.1 (b03caac1) (@vinhnx)
- Update README with enhanced TUI features and fix typo (54cbd08f) (@vinhnx)
- Update documentation for v0.12.0 with enhanced TUI features (102627cc) (@vinhnx)
- Enhance README and core library documentation for VT Code (2713a9a7) (@vinhnx)
- Refine documentation in core library for VT Code (fde12ad7) (@vinhnx)
- Update TODO.md and enhance tool permission prompts (bb3d228c) (@vinhnx)
- Enhance TODO.md with viewport changes for responsive UI (d6a8921d) (@vinhnx)
- Add RELEASES.md with detailed release notes (05862f3d) (@vinhnx)
- Update README and TODO for recent enhancements and cleanup (ad61a42c) (@vinhnx)
- Update README and vtcode.toml to reflect new LLM providers and configuration changes (ac4541ed) (@vinhnx)
- Update GitHub Packages publishing instructions and improve error messages (06d0350a) (@vinhnx)
- Add refactor proposal (99b5a23e) (@vinhnx)
- Outline reusable vtcode components (1f00cf89) (@vinhnx)
- Add crate category hashtags (7559539e) (@vinhnx)
- Stage training plan from kimi-dev research (d34a5055) (@vinhnx)
- Update AGENTS.md for clarity and structure (f340ca63) (@vinhnguyenxuan-ct)
- Add lifecycle hooks guide (53cfdaac) (@vinhnx)
- Update user guide and changelog with quick access shortcuts and enhancements (818b9530) (@vinhnx)
- Update user guide and changelog with quick access shortcuts and enhancements (b9bfa94e) (@vinhnx)
- Clean up vtcode_docs_map.md and remove unnecessary newlines in generate-agent-file.md (82a37ec3) (@vinhnx)
- Add asset synchronization guide for managing embedded assets in vtcode-core (58f253cf) (@vinhnx)
- Update README and zed-acp documentation to include package.id requirement (ceb05a65) (@vinhnx)
- Update zed-acp documentation to emphasize required schema_version in extension manifest (4e62eeb1) (@vinhnx)
- Update zed-acp documentation to clarify top-level metadata requirements in extension manifest (f37b6aa3) (@vinhnx)
- Update tool-policy and extension files to remove quotes from schema_version (fc64eabc) (@vinhnx)
- Update homebrew installation to use core tap (d153faf7) (@chenrui333)
- Add troubleshooting section for development installation (30e74b1c) (@vinhnx)
- Update Step 2 completion status and add SDK examples (7d04f7ec) (@vinhnx)
- Add Step 6 integration testing guide and test scenarios (680ae3c1) (@vinhnx)
- Complete 9-step MCP code execution roadmap with Steps 8-9 designs (2fe94d27) (@vinhnx)
- Add comprehensive MCP implementation status report (4f8fb48b) (@vinhnx)
- Add MCP quick reference guide for fast lookup (05485d38) (@vinhnx)
- Add agent prompt optimization summary (af54338c) (@vinhnx)
- Add comprehensive tool configuration status document (48da94ab) (@vinhnx)
- Add ACP implementation completion summary (1f0f0bfa) (@vinhnx)
- Add ACP next steps and release checklist (aa9930f0) (@vinhnx)
- Update ACP implementation summary and usage patterns (91f6cda1) (@vinhnx)
- Add comprehensive v0.43.0 release summary (6a7ecb88) (@vinhnx)
- Add release action checklist for v0.43.0 (8ebb5914) (@vinhnx)
- Add release readiness confirmation document (b14740db) (@vinhnx)
- Update STATUS with improvements session results (8365c391) (@vinhnx)
- Add comprehensive final improvements summary (ae04b9d1) (@vinhnx)
- Add file listing output behavior pattern to AGENTS.md (7d27dec7) (@vinhnx)
- Update installation guides with CDN caching troubleshooting and fix details (2d421f67) (@vinhnx)
- Add styling implementation completion status (c59e6869) (@vinhnx)
- Add styling implementation completion status (740119ca) (@vinhnx)
- Add phase 1 completion summary - all criteria met (aa1bb06d) (@vinhnx)
- Add session summary for phase 1 styling integration completion (63ae9b91) (@vinhnx)
- Add Phase 2 planning and implementation guides for advanced styling features (8152371a) (@vinhnx)
- Add comprehensive styling documentation index (505bba81) (@vinhnx)
- Add VT Code execution policy documentation and update command validation (8050127a) (@vinhnx)
- Update Phase 2 progress - Step 4 complete with all input methods migrated (0290d900) (@vinhnx)
- Add truncation audit and remove unused terminal output line-limit constants (8d932f15) (@vinhnx)
- Clarify run_terminal_cmd usage for git, cargo, and one-off shell commands (1a9dfff1) (@vinhnx)
- Reorganize root-level docs into docs/ subdirectories per AGENTS.md (2bbb85e5) (@vinhnx)
- Update AGENTS.md with comprehensive agent guide and tool usage guidelines (84be1513) (@vinhnx)
- Add embedded assets management guide and pre-commit hook (d7d0e7b9) (@vinhnx)
- Update PTY command execution improvements documentation (db479961) (@vinhnx)
- Add comprehensive PTY fix summary with problem analysis and solution validation (f6dae7ef) (@vinhnx)
- Add PTY fix outcome report with validation and impact assessment (1579f7c1) (@vinhnx)
- Add comprehensive PTY shell initialization fix guide (164f8775) (@vinhnx)
- Add comprehensive PTY fix outcome report with complete analysis (3b251f52) (@vinhnx)
- Fix misleading comment for RUN_PTY_CMD constant (cea0bc2e) (@vinhnx)
- Add implementation completion summary for System Prompt v3 (81908d61) (@vinhnx)
- Add External Editor Configuration to docs index (b50c0494) (@vinhnx)
- Add Ratatui FAQ-based TUI best practices guides (463c8ea3) (@vinhnx)
- Add Ratatui FAQ integration summary document (d617f6a9) (@vinhnx)
- Add async improvements documentation (3cd3304e) (@vinhnx)
- Add comprehensive Ratatui improvements summary (e498e22b) (@vinhnx)
- Update and expand documentation across various topics, add a new MCP diagnostic guide, and adjust project configurations and dependencies. (67278d19) (@vinhnx)
- Complete documentation migration - consolidate and organize (4fd42b50) (@vinhnx)
- Add comprehensive INDEX.md for navigation (aa7d7545) (@vinhnx)
- Link MCP module docs and add team guide (139a035f) (@vinhnx)
- Add team announcement document (bebe2099) (@vinhnx)
- Add executive summary document (245db406) (@vinhnx)
- Add implementation guides for immediate team use (aa542a1c) (@vinhnx)
- Add master index - 00_START_HERE.md (7b17aef8) (@vinhnx)
- Add team communication kit - ready-to-use materials (32dd9b0b) (@vinhnx)
- Add lessons learned - project retrospective (46dc2637) (@vinhnx)
- Integrate DEPLOYMENT_GUIDE and update INDEX navigation (0967a1b1) (@vinhnx)
- Update changes summary with comprehensive error handling improvements (db34e1a3) (@vinhnx)
- Clarify spinner cleanup implementation details (ff666e94) (@vinhnx)
- Add comprehensive optimization report (7623ea91) (@vinhnx)
- Add complete skill tool fix summary (c69c1b63) (@vinhnx)
- Replace all cargo nextest references with cargo test across documentation and agent rules (2161dbce) (@vinhnx)
- Add guidelines for git operations in AGENTS.md and update LLM provider configuration in vtcode.toml (609050a5) (@vinhnx)
- Update installation instructions and scripts for npm package (fa44286b) (@vinhnx)
- Add a comprehensive list of new features, bug fixes, and performance improvements to the project TODO list. (a06bd793) (@vinhnx)
- Update README with new sections for Keyboard Shortcuts and macOS Alt Shortcut Troubleshooting; refine TODO list entries for clarity and consistency. (e39f959b) (@vinhnx)
- Streamline TODO by removing verbose system skill enumeration and adding a focused task. (a0b85550) (@vinhnx)
- Add Phase 3 implementation status tracker (333a64b9) (@vinhnx)
- Add comprehensive documentation for A2A Protocol implementation (adaec652) (@vinhnx)
- Update implementation summary and configuration for file search and system prompt enhancements (b90d70a6) (@vinhnx)
- Add Phase 3 extension integration planning and file search documentation (b8117c3b) (@vinhnx)
- Add verification summary for homebrew release fixes (d769444e) (@vinhnx)
- Add guide for setting up custom homebrew tap repository (97ee2ca9) (@vinhnx)
- Explain why homebrew updates stopped and root cause analysis (da0fe91b) (@vinhnx)
- Add comprehensive homebrew fix summary with all solutions applied (c98abc61) (@vinhnx)
- Add actionable next steps for completing homebrew distribution setup (d79b416b) (@vinhnx)
- Add Homebrew verification checklist - release automation complete (ab303576) (@vinhnx)
- Update Homebrew documentation - simplified architecture (fa14d33b) (@vinhnx)
- Add Jekyll config and documentation index (359e786b) (@vinhnx)
- Add HTML landing page for GitHub Pages (62d141cf) (@vinhnx)
- Remove HTML index, use Jekyll markdown (7d0c8e0e) (@vinhnx)
- Add .nojekyll to bypass Jekyll processing (d5714c63) (@vinhnx)
- Add native installer implementation status report (49feca6c) (@vinhnx)
- Add release v0.58.6 and installer test guide (ac510a8d) (@vinhnx)
- Deployment complete - v0.58.6 release ready (e0edc6dc) (@vinhnx)
- Add native installer readme - central documentation hub (f54b14fc) (@vinhnx)
- Add release monitoring guide with auto-install instructions (b990d689) (@vinhnx)
- Add quick reference for monitoring and auto-install (ece5dff5) (@vinhnx)
- Add instructions for manually triggering release build (cc3dd6fc) (@vinhnx)
- Add Windows build fixes documentation (2eb961d5) (@vinhnx)
- Reorganize documentation and integrate skill tools into registry (efa5af44) (@vinhnx)
- Update ACP V2 Migration Guide for improved clarity and formatting (b36076e9) (@vinhnx)
- Add a new document detailing strategies for reducing Anthropic latency and link it from the Anthropic API overview. (01803299) (@vinhnx)
- Add note to check amp in vscode session (2eff5bdb) (@vinhnx)
- Add comprehensive AgentSkills support section to README (db13439f) (@vinhnx)
- Center align VT Code GIF in README (a09348eb) (@vinhnx)
- Update AGENTS.md for improved clarity and formatting (775a0db7) (@vinhnx)
- Add compliance testing section and request object for Open Responses (ee2bcf14) (@vinhnx)
- Add behavioral guidelines to reduce common LLM coding mistakes (75ca745d) (@vinhnx)
- Update skills documentation and add migration script for legacy skills (3742aa8b) (@vinhnx)
- Add task summaries feature to agent teams and enhance subagent matching logic (5fa919c1) (@vinhnx)
- Update documentation and improve clarity on execution plans, architectural invariants, and quality scores (27c61ef7) (@vinhnx)
- Update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- Update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- Add a guide for adding new models to AGENTS.md. (8648b584) (@vinhnx)
- Update TODO.md with additional PTY truncate display information and test references (cd3a3850) (@vinhnx)
- Update TODO.md with new tasks and references (049b6cae) (@vinhnx)
- Update TODO.md with examples and improve TUI display for truncated outputs (269395b1) (@vinhnx)
- Update TODO.md with critical release notes and improve terminal title (08d978a5) (@vinhnx)
- Update documentation map with enhanced topics and user questions (e7dc9369) (@vinhnx)
- Add final notes and principles for coding agents in AGENTS.md (9d2e061c) (@vinhnx)
- Update documentation structure and add Development Setup guide (7b38475a) (@vinhnx)
- Update README and authentication guide to include GitHub Copilot support (2144d1d6) (@vinhnx)
- Add Rust style guidelines and update Clippy disallowed methods (5bedfa98) (@vinhnx)
- Clarify Zed ACP local development setup (266a146c) (@vinhnx)
- Docs: update installation instructions for Homebrew and add troubleshooting note (0cb1c08d) (@vinhnx)
- Update architecture and contributing guidelines; add extension boundaries document (d17e267c) (@vinhnx)
- Update README and scripts for crate publishing process (92c4275d) (@vinhnx)
- Fix rustdoc warnings in vtcode-core (cf6887fd) (@vinhnx)
- Add Codex and debug startup release notes (9d24fb99) (@vinhnx)
- Update AGENTS.md and TODO.md for clarity on development workflow (bd4f0647) (@vinhnx)
- Remove broken copy/yank text issue description from TODO.md (29ad997b) (@vinhnx)
- Update LLM provider integration instructions for clarity (a3f6166b) (@vinhnx)
- Update README with Xiaomi MiMo V2.5 promotion details and add promotional image (786fa660) (@vinhnx)
- Add validated model pool guidance (278975c1) (@lucaszhu-hue)
- Document calculate_sha256 return format and use cases (c17ecad0) (@vinhnx)
- Update contributing guide for AI and human contributors (bca21a97) (@vinhnx)
- Enhance README with detailed command usage and workflows (62b2a576) (@vinhnx)
- Add sponsorship section with links and current sponsors (46b40322) (@vinhnx)
- Update feature and provider tables for clarity and consistency (1bc46f46) (@vinhnx)
- Add allocation optimization plan to eliminate redundant heap allocations (acc0a4ba) (@vinhnx)
- Add AGENTS.md files for various VT Code components (d56ab044) (@vinhnx)
- Enhance agent and subagent documentation with primary agent details (4cb84e23) (@vinhnx)
- Add screenshot spotlight section with GIF and image table (2b2252a0) (@vinhnx)
- Enhance ast-grep documentation in AGENTS.md and TODO.md; update system prompt to prefer ast-grep for code-shape queries (50afb69a) (@vinhnx)
- Remove Homebrew installation instruction and update provider link text (db9e269c) (@vinhnx)
- Update README to replace tables with bullet points for features and documentation (2e9081a3) (@vinhnx)
#### Features
- Enhance spinner functionality for async operations and improve user input handling (81a8e9ce) (@vinhnx)
- Add support for x-ai/grok-4-fast:free in openrouter models (a75b552b) (@vinhnx)
- Add support for x-ai/grok-4-fast:free model and update onboarding configuration (bf635e8a) (@vinhnx)
- Display reasoning traces in chat repl (44ee61cd) (@vinhnx)
- Drive chat runloop through iocraft tui (b2c21c27) (@vinhnx)
- Render spinners via iocraft placeholder (5e683d52) (@vinhnx)
- Overhaul ratatui layout and input (16ba8497) (@vinhnx)
- Improve scrolling and spinner behavior (81d763af) (@vinhnx)
- Add npm package support and update README for installation instructions (f55851f9) (@vinhnx)
- Add demo GIF and update demo settings for improved display (01b29273) (@vinhnx)
- Integrate GPT-5-Codex model into OpenAI provider and update Response API (de517db4) (@vinhnx)
- Add tool output mode configuration and enhance rendering (d26b163a) (@vinhnx)
- Add ui_surface configuration option for chat interface rendering (8e24948e) (@vinhnx)
- Enhance MCP integration and configuration management (4f171a53) (@vinhnx)
- Surface update notice in welcome banner (ea05c6f6) (@vinhnx)
- Surface update notice in welcome banner (3f20b721) (@vinhnx)
- Add Claude Sonnet 4.5 model support (76a19aee) (@vinhnx)
- Compact tool stdout rendering (4caf3b41) (@vinhnx)
- Add slash command palettes (4384db6f) (@vinhnx)
- Show write_file diff preview (d1921c5c) (@vinhnx)
- Add explicit push commands and update npm commit message in release script (4ea412e7) (@vinhnx)
- Show write_file diff preview (4a5fc971) (@vinhnx)
- Add explicit push commands and update npm commit message in release script (6c9194b4) (@vinhnx)
- Surface controls in onboarding (0bd8f410) (@vinhnx)
- Polish onboarding hints (8e92b1e7) (@vinhnx)
- Support gpt-5-codex tool handling (0a171b75) (@vinhnx)
- Add Moonshot provider and refresh model catalog (95973d8a) (@vinhnx)
- Add new tools and update policies in tool-policy.json; enhance argument parsing in text_tools.rs; update Cargo.toml for documentation (de6df449) (@vinhnx)
- Implement structured Git diff tool (60105025) (@vinhnx)
- Enhance tool execution with cancellation support and improve function declarations (efffb16b) (@vinhnx)
- Add support for GPT-OSS models and integrate Harmony tokenization (a3f123d7) (@vinhnx)
- Preserve comments when saving configuration (dfeac418) (@vinhnguyenxuan-ct)
- Add generate-agent-file prompt and update configuration path (4133db5a) (@vinhnguyenxuan-ct)
- Enhance tool approval process and update inline list selection (a66e44d7) (@vinhnx)
- Add mcp-types package with source and checksum (f84d869d) (@vinhnx)
- Extract vtcode-indexer crate (af14b73a) (@vinhnx)
- Enhance output rendering with syntax highlighting for JSON, XML, and YAML (3335d6f1) (@vinhnx)
- Add mind map screenshot to documentation (9298125c) (@vinhnx)
- Add mind map screenshot to documentation (3ecf9096) (@vinhnx)
- Update screenshot assets and remove obsolete image (54f9e2b3) (@vinhnx)
- Add VT Code VHS showcase and demo files (55f5689c) (@vinhnx)
- Add demo section with updated demo GIF in README (73d440af) (@vinhnx)
- Add comprehensive plan for open sourcing VTCode core components (51acb839) (@vinhnx)
- Run doctests separately in publish_extracted_crates.sh (bd1e070a) (@vinhnx)
- Add changelog generation from commits in release script (7c4c6a8e) (@vinhnx)
- Add comprehensive security audit and model documentation (9783e2e5) (@vinhnx)
- Implement comprehensive security documentation and fixes (99b07b79) (@vinhnx)
- Add initial package.json for VTCode Companion extension (0e69383e) (@vinhnx)
- Add initial files for VTCode Companion extension including README, LICENSE, CHANGELOG, and esbuild configuration (a64f8749) (@vinhnx)
- Add VSCode extension publishing support to release script (83bd3283) (@vinhnx)
- Add IDE integration and troubleshooting guides to documentation (05d55df6) (@vinhnx)
- Add synchronous fetching of LMStudio models and improve model selection (6c4dcf42) (@vinhnx)
- Enhance model picker and dynamic model fetching (bfd055bf) (@vinhnx)
- Add simple GitHub Pages workflow for /docs (eebca2ac) (@vinhnx)
- Implement file tree structure for file navigation (1ca5437f) (@vinhnx)
- Update README and documentation for Cursor and Windsurf support (698bb8c1) (@vinhnx)
- Update custom prompt command syntax from /prompts to /prompt (3475ae8e) (@vinhnx)
- Implement file tree structure for file navigation (cb81b543) (@vinhnx)
- Update README and documentation for Cursor and Windsurf support (08308464) (@vinhnx)
- Update custom prompt command syntax from /prompts to /prompt (c4ce81e4) (@vinhnx)
- Add MiniMax provider integration and related constants (5a5c8663) (@vinhnx)
- Add templates for agent file generation and VT Code session initiation (58be3330) (@vinhnx)
- Increase max_tool_loops to 100 and add workspace config refresh functionality (ce02ac1a) (@vinhnx)
- Add embedded asset management for prompts and documentation (4f28ae7a) (@vinhnx)
- Add asset synchronization script for managing embedded assets (50870343) (@vinhnx)
- Implement task plan management in TUI session (4b577ab0) (@vinhnx)
- Enhance command execution policies and UI interactions (d655c2f3) (@vinhnx)
- Enhance tool policy and add time conversion functions (ba190655) (@vinhnx)
- Enhance glob pattern matching to support question mark wildcard (3c2e7ed3) (@vinhnx)
- Add clear screen command to session and implement related functionality (b75df345) (@vinhnx)
- Add cross-compilation configuration and documentation (3c520021) (@vinhnx)
- Add initial implementation of VT Code Zed extension with icons, themes, and logging commands (105e457b) (@vinhnx)
- Disable Docker usage by default in build script and update Cross.toml comments (1c843ab2) (@vinhnx)
- Replace tempfile with assert_fs for improved temporary directory handling (adae675c) (@vinhnx)
- Add experimental smart summarization feature for conversation compression (4cfcdc89) (@vinhnx)
- Add VTCode Chat extension with MCP integration (f31e348d) (@vinhnx)
- Add workspace trust request functionality (1258e68e) (@vinhnx)
- Enhance workspace trust and automation features (1abd62f9) (@vinhnx)
- Add Kimi K2 Thinking model support and update Moonshot provider logic (26048b47) (@vinhnx)
- Update vtcode.toml configuration for new model provider and context optimization features (a228b8b9) (@vinhnx)
- Add configurable LLM generation parameters in vtcode.toml (b937c2c2) (@vinhnx)
- Implement NPM package publishing for VT Code (18a30f51) (@vinhnx)
- Remove deprecated tool and add test_tool to policy (69105673) (@vinhnx)
- Add OpenRouter Interleaved Thinking Implementation Plan and Quick Reference (723eccb4) (@vinhnx)
- Implement configurable MCP initialization and tool execution timeouts (2756c71f) (@vinhnx)
- Add comprehensive timeout implementation summary and configuration details (45ec34d0) (@vinhnx)
- Step 2 - Code executor with SDK generation and IPC (61058cf6) (@vinhnx)
- Step 2 Phase 2 - IPC handler integration for tool invocation (de0007b6) (@vinhnx)
- Implement all 5 MCP code execution steps from Anthropic recommendations (c1143740) (@vinhnx)
- Step 7 - Observability & Metrics system for MCP execution (3d00a080) (@vinhnx)
- Step 8 - Implement tool versioning and compatibility checking (b17c1823) (@vinhnx)
- Implement tool approval dialog with enhanced UX and risk assessment (f8c648bf) (@vinhnx)
- Implement Agent Communication Protocol (ACP) integration (e8171ae5) (@vinhnx)
- Implement styling refactor - centralize color palettes and style helpers (81fb334a) (@vinhnx)
- Integrate anstyle-parse for ANSI escape sequence handling (e3131002) (@vinhnx)
- Complete phase 1 anstyle integration - effects and background colors (dc399246) (@vinhnx)
- Add theme_parser module for Git/LS_COLORS configuration parsing (a7dd9657) (@vinhnx)
- Add Styling Quick Start Guide and Refactor Completion Report (7d343f5d) (@vinhnx)
- Add Git color configuration support and theme management (e9eb73e2) (@vinhnx)
- Implement permission system with command resolution, audit logging, and caching (ebfd8da6) (@vinhnx)
- Enhance command execution with additional PATH entries and environment variable handling (7c4b1267) (@vinhnx)
- Implement token-based truncation for tool output rendering (00272d60) (@vinhnx)
- Introduce Web Fetch tool with security configurations (569e65e2) (@vinhnx)
- Refactor loop hang detection for improved accuracy and user experience (bb37e15a) (@vinhnx)
- Add automatic installation and management for ripgrep dependency (74ec2b47) (@vinhnx)
- Implement token-based truncation for tool outputs and update configuration (57f20038) (@vinhnx)
- Enhance diff display with full-width backgrounds and improve terminal command visibility (cdfd3b92) (@vinhnx)
- Implement interactive tree UI for file structure visualization (9124a7e9) (@vinhnx)
- Add Bash tool and remove non-existent run_terminal_cmd references (8e342d57) (@vinhnx)
- VT Code System Prompt v3 - Context Optimized Implementation (5df555be) (@vinhnx)
- Update tool policies, add setup script, and enhance README with configuration details (27990459) (@vinhnx)
- Add external editor integration with TUI suspension, alternate screen handling, and stability improvements. (c113eec8) (@vinhnx)
- Add default editor fallback (vi on Unix, notepad on Windows) when EDITOR/VISUAL not set (b80b1231) (@vinhnx)
- Add new tools to tool-policy and update permissions for fetch and time providers (d18c9a83) (@vinhnx)
- Refactor install script, rename `run_pty_cmd` to `run_terminal_cmd`, and update installation instructions for Homebrew and NPM. (f53e126e) (@vinhnx)
- Add comprehensive ANSI escape sequence documentation and a new core utility module for ANSI codes. (b47807c1) (@vinhnx)
- Implement animated thinking spinner for user input submission (be2904cd) (@vinhnx)
- Integrate production-grade tool improvements system (5ad45963) (@vinhnx)
- Enhance tool policy with pre-approval allowlist, improve file operation error messages, and refine tool declarations. (708b8c1e) (@vinhnx)
- Parse and display friendly error messages from Anthropic API responses (d1d978a1) (@vinhnx)
- Add diff suppression logic and constants for large changes (1dc090d5) (@vinhnx)
- Update tool policies and improve code structure with dead code allowances (733e2575) (@vinhnx)
- Add multi-stage Dockerfile for building and running vtcode (0d1aa194) (@leonj1)
- Implement tool call safety validation and execution tracking (3fb3bd30) (@vinhnx)
- Update tool policy and enhance loop detection functionality (13475bdc) (@vinhnx)
- Add search_replace tool and enhance file operations (6e7a4f3c) (@vinhnx)
- Enhance timeout configuration and adaptive timeout handling (6358dcc8) (@vinhnx)
- Enhance grep result optimization and tool registration (69cf81e0) (@vinhnx)
- Add new dependencies and improve error handling in main (e13433a0) (@vinhnx)
- Introduce reinforcement learning optimization and enhance configuration (77c3103a) (@vinhnx)
- Enhance agent logging and improve markdown rendering (0c4cd820) (@vinhnx)
- Implement streaming failure management and cooldown mechanism (aaa7b64f) (@vinhnx)
- Implement timeout management for streaming and generation requests (aff6c50a) (@vinhnx)
- Enhance loop detection and rate limiting in tool execution (6a355d77) (@vinhnx)
- Improve HTTP client pool handling and enhance caching middleware (0559fc7d) (@vinhnx)
- Implement idle turn detection and management in task execution (bf9ecde3) (@vinhnx)
- Improve tracing initialization and error handling in main (b0433656) (@vinhnx)
- Enhance logging and error handling in orchestrator and agent components (198d31ca) (@vinhnx)
- Enhance tool execution with rate limiting and workspace management (3fbd26b2) (@vinhnx)
- Implement timeout warning management for tool execution (437b456c) (@vinhnx)
- Enhance response output with reasoning traces (70617716) (@vinhnx)
- Update OpenAI provider to support GPT-5.2 and enhance reasoning options (338cda79) (@vinhnx)
- Implement rate limiting for tool calls and add Nemotron-3-Nano model support (cfdf7587) (@vinhnx)
- Add Gemini 3 Flash Preview model and update configurations (1f14429e) (@vinhnx)
- Enhance line ending handling in patch operations and tests (b3734419) (@vinhnx)
- Implement shell command policy checking with regex and glob patterns and add new metadata fields to skill definitions. (9d023c15) (@vinhnx)
- Enhance skill definitions with new metadata fields, improve LLM provider support, and refine TUI components. (fe84eccd) (@vinhnx)
- Update tool policy to prompt for file creation and execution, enhance session state management, and adjust LLM provider settings in configuration. (91b0d16e) (@vinhnx)
- Add success indicators to renderer on exit commands and session end (bef872e6) (@vinhnx)
- Add code reviewer and commit message generator skills (86a30ce9) (@vinhnx)
- Implement circuit breaker pattern for MCP client failures and optimize tool inventory management (7e579111) (@vinhnx)
- Implement per-tool rate limiting and refactor agent tool execution state management. (874a533a) (@vinhnx)
- Add full-auto mode support and update tool policies for improved automation (a814298b) (@vinhnx)
- Add context-aware prompt enrichment (vibe coding) with new context modules and wizard modal interaction events. (ea4798e0) (@vinhnx)
- Remove reinforcement learning and optimization modules and configurations. (358ba06a) (@vinhnx)
- Extract anthropic config, reduce configuration complexity, document experimental features (2e3aa961) (@vinhnx)
- Enhance tool execution policy with granular user confirmation, auto-acceptance, and feedback capabilities. (8eda995d) (@vinhnx)
- Implement Agent Skills specification by adding `compatibility` and `metadata` fields to skill manifests, updating `allowed-tools` to a space-delimited string, and clarifying skill loading behavior. (26fd647b) (@vinhnx)
- Enhance skill validation and file reference checks for Agent Skills compliance (bedb8cea) (@vinhnx)
- Implement agent task retry with exponential backoff and render tool follow-up prompts. (6e54c684) (@vinhnx)
- Enable manual versioned builds in the release workflow and significantly enhance the install script with improved dependency/platform detection, asset verification, and a cargo fallback. (cc9e5502) (@vinhnx)
- Add npm publishing troubleshooting guide and authentication setup script, and automate binary stub creation in the release process. (885cecf0) (@vinhnx)
- Add Linux build and release support, fix npm publish, and improve release asset uploads and install script error handling. (e9a7f51d) (@vinhnx)
- Improve release process by adding Linux build automation and related documentation. (b8d85725) (@vinhnx)
- Implement pre-flight LLM request and tool definition validation, and ensure `mark_tool_loop_limit_hit` is idempotent. (1424aa15) (@vinhnx)
- Add HITL notification bell configuration and implement terminal bell notification for approvals (09bb2a4f) (@vinhnx)
- Add plan phase management and update tool registry for planning mode (d0db1619) (@vinhnx)
- Implement TUI-aware tool approval prompts and human-in-the-loop notification bell. (0d3214bf) (@vinhnx)
- Enhance tool execution error handling and implement planning mode warnings (282ae823) (@vinhnx)
- Implement XDG Base Directory Specification for configuration and data storage (2fd4c1e8) (@vinhnx)
- Enhance list rendering with highlight symbol and repeat option (042bd041) (@vinhnx)
- Add better panic handling with debug mode support (3d9c9925) (@vinhnx)
- Implement centralized panic handling for TUI applications (9e7d67a0) (@vinhnx)
- Create ratatui widget foundation (3af9f457) (@vinhnx)
- Add buffer-based widgets for input, modal, and slash (381307bb) (@vinhnx)
- Integrate SessionWidget into main render function (0ffea35b) (@vinhnx)
- Enhance session logging functionality and update default model (f37f4d36) (@vinhnx)
- Introduce a TUI config palette, refactor rendering logic, and enable dynamic theme application. (b5514792) (@vinhnx)
- Implement search and filter functionality for the TUI configuration palette. (2d845515) (@vinhnx)
- Add new model constants for grok-4-1-fast and grok-code-fast-1 (cdbe1e9e) (@vinhnx)
- Introduce advanced LLM parameters, add default implementations for LLMRequest, Message, and ToolChoice, and remove nextest.toml. (74b84679) (@vinhnx)
- Add keyboard protocol configuration and documentation for enhanced keyboard event handling (dc9dc5d9) (@vinhnx)
- Introduce agent giving-up reasoning detection and constructive responses, and set `execute_code` tool policy to prompt. (49b36236) (@vinhnx)
- Include Claude agent configurations and GitHub workflows in version control, and update existing agent definitions, skills, commands, hooks, and CI/CD configurations. (c6dc5186) (@vinhnx)
- Add Z.AI GLM-4.7 model to models.json and update constants (01777c80) (@vinhnx)
- Update tool policies and add new Grok models to configuration (8dd3bdef) (@vinhnx)
- Enhance OpenAI responses handling with tool call parsing and sampling parameters (3fd43234) (@vinhnx)
- Add missing OpenRouter model entries and update reasoning handling (e067daef) (@vinhnx)
- Update model references and configuration for Z.AI GLM-4.7 (385ea5f0) (@vinhnx)
- Expand Hugging Face model support and update provider implementation (df6a5219) (@vinhnx)
- Update model provider to OpenAI and enhance Responses API handling (928239b0) (@vinhnx)
- Add support for HuggingFace provider selection and models (b179a3c4) (@vinhnx)
- Add Hugging Face provider support and update configuration (82a1cb9a) (@vinhnx)
- Reorganize Hugging Face model identifiers and enhance Anthropic model validation (cb5a5854) (@vinhnx)
- Add MiniMax model support to the Anthropic provider and adjust its API base URL. (d8602c4c) (@vinhnx)
- Add Hugging Face integration documentation and update tool policies to include git and cargo commands while removing some mcp time-related tools. (28b0965e) (@vinhnx)
- Reimplement HuggingFace LLM provider with dedicated logic to handle its unique API behaviors and compatibility. (bb973323) (@vinhnx)
- Add new Z.AI GLM models, refine reasoning support, and update Hugging Face model naming conventions. (d5957215) (@vinhnx)
- Add code intelligence tool with LSP-like navigation features (936abcb4) (@vinhnx)
- Implement terminal setup wizard with support for multiple terminals and features, and update LLM provider models. (857d4b5b) (@vinhnx)
- Refactor and expand slash command handling with new diagnostics, skills, tools, workspace, and context commands. (13d3b062) (@vinhnx)
- Introduce `--quiet` flag and separate `stdout` for data and `stderr` for logs to improve CLI piping. (2b8f5e16) (@vinhnx)
- Introduce GLM-4.7 Novita model, prepend system prompts in HuggingFace provider, skip GLM thinking parameter, and update tool policies. (7fbacf2c) (@vinhnx)
- Improve tool input deserialization to handle quoted values, enhance `grep` path validation, and update tool policies. (4ed5d4c4) (@vinhnx)
- Introduce a centralized HTTP client factory, refactor providers to use it for consistent timeout configuration, and enhance API error parsing. (5fb79ffc) (@vinhnx)
- Integrate `TimeoutsConfig` into LLM provider HTTP clients and refactor OpenRouter error handling. (1c66819f) (@vinhnx)
- Improve terminal detection and configuration path resolution across operating systems, update LLM provider integrations, and refine agent slash commands and welcome flow. (d4947016) (@vinhnx)
- Add alias for /config command as /settings, enhance slash command descriptions, and introduce quiet mode in configuration (2ddd4059) (@vinhnx)
- Implement adaptive TUI tick rate and coalesce scroll events (4decf82f) (@vinhnx)
- Implement adaptive TUI tick rate, coalesce scroll events, and enhance session management (8bfd6a32) (@vinhnx)
- Improve error handling for create_file and update_plan methods, enhance logging for theme loading failures (29788638) (@vinhnx)
- Enhance tool execution logging and improve diff preview generation (9c4ea308) (@vinhnx)
- Enhance input history navigation and improve session input handling (635e5d6a) (@vinhnx)
- Enhance planning tool with quality validation and detailed descriptions for task phases (e96e082b) (@vinhnx)
- Add agent option to CLI for temporary model override (dee29346) (@vinhnx)
- Optimize ANSI syntax highlighting in diff renderer for improved performance (4ea8db1b) (@vinhnx)
- Enhance documentation and prompts for clarity, consistency, and performance improvements (23bb5334) (@vinhnx)
- Implement session forking with custom session ID support (ca3b4824) (@vinhnx)
- Add session resumption functionality and update related actions (00b43e11) (@vinhnx)
- Enhance session resume/fork logic and improve conversation history display during session startup. (661ca205) (@vinhnx)
- Enhance textual tool call parsing, pre-validate arguments, and refine tool failure detection to improve agent robustness. (4e52efeb) (@vinhnx)
- Introduce dynamic system prompt enhancements including temporal context and working directory awareness, along with refined tool usage guidelines for improved agent performance. (9ca87af3) (@vinhnx)
- Update default agent configuration to HuggingFace and refine tool schemas and prompt generation logic. (323bbb21) (@vinhnx)
- Implement LSP client and manager with agent slash commands, and add LLM provider caching tests. (570ec3bd) (@vinhnx)
- Add `--skip-release` option and enhance GitHub release verification logic with CI environment detection. (dbec0942) (@vinhnx)
- Implement timed double Ctrl+C for agent exit, deferring shutdown, and update the default model. (c82b4cb2) (@vinhnx)
- Implement PTY session termination on Ctrl+C cancellation with debounced signal handling and status line feedback. (2eb2709b) (@vinhnx)
- Add `mcp::fetch` and `mcp::time` tools, simplify LSP client message handling, and remove outdated agent system analysis from TODO documentation. (a048cf38) (@vinhnx)
- Calculate context usage from history and add a final pre-request safety check after trimming. (15846421) (@vinhnx)
- Introduce context summarization with adaptive trimming integration and new `Summarize` retention choice. (7b0590c8) (@vinhnx)
- Add `LoadSkillTool` for progressive skill instruction loading, enhance skill context with path storage, and integrate skill discovery into agent setup. (db8e0dad) (@vinhnx)
- Add `ListSkillsTool` for programmatic skill discovery, replacing direct skill prompt integration. (ddf86ffc) (@vinhnx)
- Implement on-demand skill loading with `LoadSkillTool` and `LoadSkillResourceTool`, and enable skill restoration from previous sessions. (1f84ffbc) (@vinhnx)
- Add new Ollama cloud models and update reasoning model detection. (b5be9384) (@vinhnx)
- Implement `Tool` trait for `CliToolBridge` and integrate skill-based tool registration with `ToolRegistry`. (7e557d5c) (@vinhnx)
- Introduce skill varieties and enhance skill listing/loading with filtering and dormant tool support (5bf77f9c) (@vinhnx)
- Implement lazy-loading and tiered disclosure for agent skills and tools, reducing default available tools and updating system prompts. (598d0811) (@vinhnx)
- Refine tool policies and skill loading for lazy-loaded capabilities, updating system prompts to reflect on-demand activation. (c3ae7720) (@vinhnx)
- Reimplement skill management with a new skill model and dedicated modules. (eb991ddf) (@vinhnx)
- Add support for loading skill `references/` and `assets/` directories and introduce `ResourceType::Asset` (ac20bd7a) (@vinhnx)
- Introduce `EnhancedSkillLoader` and `EnhancedSkill` for unified skill and tool management, and refactor skill discovery results across the agent and CLI. (4395972b) (@vinhnx)
- Add GitHub account switching and cleanup functionality in release script (8d88b783) (@vinhnx)
- Refactor tool permission handling for TUI-only mode and update default LLM provider configuration. (92b6deb4) (@vinhnx)
- Remove the `plan` tool and associated components, and update related tool and skill management logic. (efb2072c) (@vinhnx)
- Add async method to InlineSession for receiving next event (1ed70c7d) (@vinhnx)
- Implement Agent2Agent (A2A) Protocol support (Phase 1 & 2) (b14b8e15) (@vinhnx)
- Implement full SSE streaming support (Phase 3.1) (30d44390) (@vinhnx)
- Add webhook notifier for push notifications (Phase 3.2 partial) (495cc9eb) (@vinhnx)
- Finish push notification config storage and RPC wiring\n\n- Add webhook config storage to TaskManager (set/get/remove) with SSRF validation\n- Wire JSON-RPC handlers for pushNotificationConfig set/get\n- Fix server dispatch and imports\n- All A2A tests pass (37/37) including server + webhook (e482c2ea) (@vinhnx)
- Trigger webhooks on streaming events\n\n- Add webhook_notifier to server state and wire into streaming pipeline\n- Fire webhooks for status updates and messages when broadcasted\n- Fix SSRF-safe config retrieval and avoid Option to_string() error\n- Clean up unused tracing import in webhook module\n- Tests: all A2A suites pass (37/37) (407b88f8) (@vinhnx)
- Add A2A client with streaming support\n\n- New A2aClient for discovery, task ops, push config, and streaming\n- SSE client parses streaming events without extra deps\n- Simple incremental request IDs and HTTPS agent card fetch\n- Tests added for SSE parsing helpers\n\nTests: cargo test --package vtcode-core --lib a2a (39/39) (138ff8d6) (@vinhnx)
- Complete CLI integration and documentation\n\n- Add full A2A CLI with serve, discover, send-task, list-tasks, get-task, cancel-task commands\n- Create comprehensive CLI handlers for all A2A operations\n- Fix streaming event handling with proper pinning\n- Update server.rs Box<dyn Stream> return type for axum compatibility\n- Add completion summary document\n- All checks pass: cargo check --package vtcode-core\n\nImplements: A2A Protocol Phase 4 - CLI integration and user-facing features (292754fd) (@vinhnx)
- Enhance GitHub account handling for CI environments (6abf68d1) (@vinhnx)
- Add support for multiple workspaces and enhanced security controls (8842b8d5) (@vinhnx)
- Add toggle for terminal notifications in config (79403fa0) (@vinhnx)
- Add marketplace and plugin management commands (2bfb2412) (@vinhnx)
- Add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications (6a0fdc94) (@vinhnx)
- Implement output styles feature with customizable behavior and response formats (e2429e50) (@vinhnx)
- Add reverse search and background operation handling in TUI (a8d2bcdb) (@vinhnx)
- Add custom status line scripts and JSON input handling (9edeb27c) (@vinhnx)
- Document subagent system and built-in agents; update README and changelog (0f461988) (@vinhnx)
- Add tokio dependency and improve error handling in background task execution (476cb461) (@vinhnx)
- Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX. (1c06838b) (@vinhnx)
- Improve context manager performance with incremental stats and bolster tool security with path and command validation. (a7245390) (@vinhnx)
- Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`. (5e80dd86) (@vinhnx)
- Implement vtcode-file-search crate and integration with extensions (29764ed8) (@vinhnx)
- Introduce vtcode-process-hardening crate for enhanced security measures (cf02a8ef) (@vinhnx)
- Implement comprehensive Ollama client and progress reporting system (f781c61d) (@vinhnx)
- Update Cargo.lock and Cargo.toml with new packages and versions (67085671) (@vinhnx)
- Revise system prompts for clarity, autonomy, and Codex alignment (191148da) (@vinhnx)
- Enhance command safety module with comprehensive integration and documentation (9ec89453) (@vinhnx)
- Add move and copy file operations with error handling (5e2d89f6) (@vinhnx)
- Add keyboard protocol environment check and enhance logging (20d78b88) (@vinhnx)
- Add native installer with auto-updater module (660c7be2) (@vinhnx)
- Auto-trigger build-release workflow on GitHub release creation (25aff6dd) (@vinhnx)
- Refactor tool permission context and enhance command safety validation with new progress updates for tool execution. (4a9871ee) (@vinhnx)
- Implement parallel tool execution for agent actions and update tool registry operations to be asynchronous. (41c36f66) (@vinhnx)
- Refactor tool registry to use MCP tool index cache and update mutability of inventory and tool policy access. (56d15fa7) (@vinhnx)
- Enhance LLM provider initialization with client injection and refine prompt caching (980a1602) (@vinhnx)
- Integrate real performance optimizations into tool registry (66cc5c5c) (@vinhnx)
- Add file system watcher and performance optimization infrastructure (eedafab9) (@vinhnx)
- Enhance tool caching and UI redraw optimization (4bff5beb) (@vinhnx)
- Restore Kitty keyboard protocol support and update session handling (fa00356a) (@vinhnx)
- Add LRU cache for canonicalized paths and optimize vector allocations (0984888f) (@vinhnx)
- Implement dynamic context discovery with file spooling for large outputs (47e58f8a) (@vinhnx)
- Enhance dynamic context discovery and update configuration (f79caf24) (@vinhnx)
- Implement sandboxing configuration and policies (3e38134d) (@vinhnx)
- Implement Plan Mode for read-only exploration and planning (cab12fcd) (@vinhnx)
- Add editing modes and commands for toggling between Edit, Plan, and Agent modes (222acd22) (@vinhnx)
- Update tool policies to allow apply_patch, unified_exec, and unified_file actions (07377b04) (@vinhnx)
- Implement session limit increase prompt and safety validation enhancements (ed33761c) (@vinhnx)
- Implement plan mode tools for managing planning workflow and enhance code block indentation normalization (286cbc8e) (@vinhnx)
- Update tool policies and enhance session limit handling for tool loops (48f0894c) (@vinhnx)
- Enhance sandboxing with new documentation and environment handling (f4a10c1e) (@vinhnx)
- Add GPT-5.2 Codex model and improve code formatting across multiple files (d85983d5) (@vinhnx)
- Enhance agent behavior configuration with Codex-inspired patterns and update tool response truncation settings (817ec823) (@vinhnx)
- Update tool policies and agent configuration for improved execution control and user confirmation (bea42960) (@vinhnx)
- Update tool policies, enhance file handling, and modify agent configuration for improved functionality (2011170c) (@vinhnx)
- Add debug logging for subagent parsing and loading; update error messages for context and segment not found (a7d9e949) (@vinhnx)
- Add Anthropic API compatibility server and documentation (b7245fff) (@vinhnx)
- Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types. (1a99ccbe) (@vinhnx)
- Add tool search configuration and integration for Anthropic provider (b4e88613) (@vinhnx)
- Add support for new Anthropic Claude 4 and 3.x models, enable new beta features, and refine reasoning parameter validation. (9f4bd9b1) (@vinhnx)
- Implement support for request-specific Anthropic beta headers and update structured output model list. (fde8052c) (@vinhnx)
- Implement request and organization IDs for LLM responses and error metadata, enhance Anthropic error handling, and add `Refusal` finish reason. (b49e0202) (@vinhnx)
- Add `thinking_budget` to `LLMRequest` and implement Anthropic extended thinking logic and validation. (dd741d01) (@vinhnx)
- Add prefill and character reinforcement options to LLMRequest, implement Anthropic-specific handling, safety screening, and leak protection. (23339341) (@vinhnx)
- Introduce coding agent settings to LLM requests to refine model behavior, implementing their application in the Anthropic provider for system prompt adjustments, prefill, message reordering, and XML document handling. (b75d3d0e) (@vinhnx)
- Implement a new evaluation framework with test cases, metrics, and report generation, and update LLM provider integrations to support it. (f1dc5d50) (@vinhnx)
- Add an empirical evaluation framework for measuring LLM performance and link it in the main README. (11f7bc12) (@vinhnx)
- Improve code block indentation normalization to handle mixed whitespace and refine markdown table rendering separators. (f64e9125) (@vinhnx)
- Introduce dedicated modules for MCP lifecycle, slash command handling, and tool dispatch, and parallelize tool batch execution. (a4d1deea) (@vinhnx)
- Enhance tool execution with circuit breakers, adaptive rate limiting, and health-based delegation, and introduce session telemetry and dynamic cache capacity management. (88e627c6) (@vinhnx)
- Implement adaptive rate limiting with priority-based scaling and integrate telemetry for tool usage tracking. (544e1ac8) (@vinhnx)
- Dynamically configure conversation message and session turn limits, and remove telemetry from interaction loop parameters. (e6b687e9) (@vinhnx)
- Update session limit messages to recommend persisting progress via artifacts like task.md/docs. (0a5a3a00) (@vinhnx)
- Implement priority-based adaptive rate limiting, tiered cache eviction, and sliding window tool health tracking. (24d02051) (@vinhnx)
- Enhance agent robustness with exponential backoff for circuit breakers, custom tool loop limits, and conversation turn limits. (ad3105bf) (@vinhnx)
- Implement autonomous loop detection with TUI warnings, integrate into agent runloop, and add project TODO documentation. (27674f6c) (@vinhnx)
- Add max_conversation_turns configuration to various components and update tests (6f4c60f4) (@vinhnx)
- Enhance context awareness with token usage tracking and context window size (211bd094) (@vinhnx)
- Enhance output spooling for read_file and unified_file with raw content extraction (a290ad68) (@vinhnx)
- Implement autonomous mode with reduced HITL prompts and update related configurations (0b5e7c5f) (@vinhnx)
- Add effort parameter for Claude Opus 4.5 to control token usage (dad1cf1a) (@vinhnx)
- Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option. (07c60d20) (@vinhnx)
- Add extended thinking configuration for Anthropic models (8fb4603a) (@vinhnx)
- Add Z.AI GLM-4.7-Flash model support and update configuration (ae9c7f85) (@vinhnx)
- Add support for image URLs in @ pattern parsing and implement vision support for LLM providers (5e649dac) (@vinhnx)
- Enhance CLI with quick start guidance and slash command notes (81eb783b) (@vinhnx)
- Add support for additional programming languages including swift in syntax highlighting and configuration (61a1a350) (@vinhnx)
- Implement batch file reading with token-efficient command transformation (a00fa4d3) (@vinhnx)
- Add adaptive logo SVGs for different color schemes (0ed49c50) (@vinhnx)
- Add harness event emitter and session persistence for enhanced logging and state management (03b65cef) (@vinhnx)
- Add subagent system with optional enablement, commands, and configuration updates (8e340fbc) (@vinhnx)
- Add nested discovery for Claude skills, enhance SKILL.md parsing with default values, and update validation rules (36359e14) (@vinhnx)
- Refine tool output styling, and enhance agent configuration (b2e65e9d) (@vinhnx)
- Refine tool output styling, and enhance agent configuration (acf563fc) (@vinhnx)
- Introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation (9ae61367) (@vinhnx)
- Introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation (75d767f2) (@vinhnx)
- Migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling (43310840) (@vinhnx)
- Migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling (d602b176) (@vinhnx)
- Add human-readable slug generator for plan file naming, update TUI header editing mode handling (e5edf836) (@vinhnx)
- Add OpenAI provider support for chat message parsing, request building, response parsing, and streaming decoder implementations (8ce1b3a9) (@vinhnx)
- Add `parse_openai_tool_calls` function to handle OpenAI tool call parsing logic (98a75f8a) (@vinhnx)
- Integrate `MCP client` with `ToolRegistry` and add functions for tool management (d94c3e9d) (@vinhnx)
- Add `ZedAgent` implementation to support session management, tool execution, and client interaction (eb33ff4c) (@vinhnx)
- Modularize tool outcome handlers into separate files (failure, success, timeout, apply) and refactor implementation for better readability and maintainability (23d951b9) (@vinhnx)
- Implement AgentRunner modularization for summarization, telemetry, tool access, and execution (fbf874c4) (@vinhnx)
- Streamline `file_ops` by removing legacy recursive search methods and enhance OpenAI provider with streaming logic (367111db) (@vinhnx)
- Add enhanced caching logic and tool execution pipeline improvements (dd82f406) (@vinhnx)
- Add UI support for modal layouts and wizard states in TUI session (fb2a5141) (@vinhnx)
- Add new logo assets and update existing images for branding consistency (e58a5f78) (@vinhnx)
- Add GitHub Actions release workflow and update release script for better error handling (4c50ba26) (@vinhnx)
- Introduce a standard Agent Client Protocol adapter and generalize ACP implementation details and tooling. (121184b5) (@vinhnx)
- Add `switch_mode` tool and update `agent-client-protocol` dependency to 0.9.3, adapting API usage. (1bdf7a7f) (@vinhnx)
- Implement session loading and mode switching, refreshing available commands on mode change and using constants for mode IDs. (9004549c) (@vinhnx)
- Implement dotfile protection with audit, backup, and guardian modules, and enhance tool registry alias resolution to prioritize LLM-visible tools. (62689a2c) (@vinhnx)
- Prevent duplicate LLM reasoning output and prioritize visible alias targets for hidden tools during lookup. (c434de61) (@vinhnx)
- Add skip_model_validation option to AnthropicConfig and update validation logic (f19fdb56) (@vinhnx)
- Implement history picker for fuzzy command search (Ctrl+R) (e529e7a0) (@vinhnx)
- Add ANSI escape sequence parsing constants and improve handling in text utilities (d5641f54) (@vinhnx)
- Add ACP authentication methods and configuration support (e51a5658) (@vinhnx)
- Implement wire API detection and version handling for Ollama (0cc27230) (@vinhnx)
- Enhance Agent Trace support with async storage and serialization improvements (e777f259) (@vinhnx)
- Enhance Turn Diff Tracker with Agent Trace support and backward compatibility (a81074bd) (@vinhnx)
- Update test commands to prefer cargo-nextest for faster execution (db00e155) (@vinhnx)
- Update imports in harness.rs and tests.rs for improved clarity (6e2b9432) (@vinhnx)
- Enhance AGENTS.md with new cargo commands and build performance tips (4df82383) (@vinhnx)
- Add signal handling for graceful termination in TUI (f4de0101) (@vinhnx)
- Add Step 3.5 Flash model and update configuration for OpenRouter (c50770ab) (@vinhnx)
- Update tool policies and enhance message handling; modify configuration for LLM provider (1c093ab7) (@vinhnx)
- Reject hooks in skill definitions and update validation logic (6a4105e9) (@vinhnx)
- Update tool policies, enhance message rendering, and modify default model configuration (436ac6cb) (@vinhnx)
- Enhance UI styling and message rendering; improve error and info message handling (fa61cea9) (@vinhnx)
- Enhance message rendering for info boxes; group consecutive info messages and improve styling (3557bd25) (@vinhnx)
- Enhance debug script to handle sccache permission errors during build and run (0beccb9e) (@vinhnx)
- Improve sccache error handling in debug script; retry without sccache on permission errors (29bff726) (@vinhnx)
- Simplify debug script by removing sccache handling; enhance markdown diff rendering (cd8cdd28) (@vinhnx)
- Enhance file output rendering to display diff content when applicable (0c05c762) (@vinhnx)
- Add Qwen3 Coder Next model with enhanced reasoning capabilities (26a65840) (@vinhnx)
- Enhance diff view with changed lines count summary and line numbers (808464e9) (@vinhnx)
- Add support for inline streaming and recent spooled output retrieval (eee866c7) (@vinhnx)
- Enhance diff rendering with summary formatting and colorization (bbed557c) (@vinhnx)
- Clarify patch input parameters and remove 'diff' alias to prevent confusion (466549b7) (@vinhnx)
- Update text deletion commands for improved line handling (7c003cae) (@vinhnx)
- Enhance agent message rendering with left padding and improved line handling (dc8025a6) (@vinhnx)
- Add git diff guidance to tool guidelines (934d723e) (@vinhnx)
- Implement shell snapshot feature to optimize command execution (e5d9d7fe) (@vinhnx)
- Enhance command safety checks for git subcommands and improve branch operation validation (ca9833f4) (@vinhnx)
- Add turn metadata support for LLM requests with git context (46a57d6d) (@vinhnx)
- Add support for inline data URLs and images in message content (90254442) (@vinhnx)
- Implement queue editing functionality and update input handling (6c0d373b) (@vinhnx)
- Enhance input handling with queue overlay and update input placeholders (8394895f) (@vinhnx)
- Enhance Plan Mode with reminders and execution prompts (ba69b139) (@vinhnx)
- Implement command blocking during running tasks and update configuration for LLM provider (da55d7cd) (@vinhnx)
- Add path utilities and normalize ASCII identifiers for improved path handling (5a888daf) (@vinhnx)
- Improve plugin validation and enhance path resolution in PTY manager (a2ea12c1) (@vinhnx)
- Integrate `tui_input` crate for enhanced input management and modal search functionality (69a24ed3) (@vinhnx)
- Add 'mono' theme and improve TUI modal search input handling. (7fd6334c) (@vinhnx)
- Introduce compile-time optimization guide and profiling script, and add general performance principles to TODO. (f1257ba9) (@vinhnx)
- Display a scroll indicator in the TUI footer and adjust status height calculation based on layout mode. (6c5efc03) (@vinhnx)
- Add delete_word_forward method to InputManager and update LayoutMode footer behavior (29e2dd09) (@vinhnx)
- Implement delete_word_forward method in Session (5936d8bc) (@vinhnx)
- Enhance crate publishing process with reliable version parsing and no-verify option (4fb5612c) (@vinhnx)
- Add wait_for_crates_io function to ensure crate availability on crates.io (8b4ac577) (@vinhnx)
- Add pty_stream module and integrate it into tool pipeline execution (17b0c9d6) (@vinhnx)
- Complete model migration, fix test failures, and enhance UI stability (32f252ec) (@vinhnx)
- Add support for MoonshotAI Kimi K2 models in ModelId (90e18ff2) (@vinhnx)
- Add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
- Add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- Add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- Implement plan mode toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- Implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- Add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- Add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- Implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- Add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- Implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- Add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- Implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- Add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- Add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- Refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (db99f4db) (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (c90f06e3) (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (83567152) (@vinhnx)
- Add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (d460c56d) (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (dc0742c0) (@vinhnx)
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (8b5b42a1) (@vinhnx)
- Add `prompt_cache_key` to OpenAI requests for improved cache locality and simplify Responses API usage logic. (45c9002e) (@vinhnx)
- Add top-level cache control to Anthropic requests, with TTL determined by breakpoint consumption. (91c0c9e4) (@vinhnx)
- Standardize MiniMax-M2.5 model identifier, promote it as the default, and update configuration defaults. (ff6dcef6) (@vinhnx)
- Introduce CI cost optimization strategies, add a new `--ci-only` release mode, and document release workflow details. (dd2f3168) (@vinhnx)
- Add prompt cache key to LLM requests and enhance unified_file tool execution diagnostics. (4073aed6) (@vinhnx)
- Refactor Ollama non-streaming request handling and add a fallback to non-streaming for initial stream failures. (30683331) (@vinhnx)
- Improve spooled tool output handling by verifying file existence and add a mechanism to suppress agent follow-up prompt detection for auto-generated prompts. (083ae71a) (@vinhnx)
- Enhance error handling and recovery mechanisms across various components (7902206c) (@vinhnx)
- Implement tool reentrancy guard to prevent recursive execution and improve panic reporting with `better-panic`. (44351bf5) (@vinhnx)
- Implement chunked reading for spooled tool outputs with improved agent messaging and update default LLM provider configuration. (b5b0c230) (@vinhnx)
- Add chunked file read spool progress tracking and refine token usage calculation for context management. (ef8f162d) (@vinhnx)
- Generate consolidated checksums.txt for releases and centralize script utilities into common.sh. (4bdb0808) (@vinhnx)
- Implement TaskTracker tool and enhance agent guards and documentation based on NL2Repo-Bench insights. (45a9a159) (@vinhnx)
- Integrate AI agent best practices into system prompts and loop detection for improved planning, root cause analysis, and uncertainty recognition. (91e5e295) (@vinhnx)
- Enhance documentation on grounding, uncertainty, and regression verification; improve loop detection guidance (064ea630) (@vinhnx)
- Enhance `AskUserChoice` with freeform input, custom labels, placeholders, and default selections. (53e0e111) (@vinhnx)
- Implement freeform text input for wizard modals, guided by system prompt and toggled by the Tab key. (9b54cdd6) (@vinhnx)
- Refine plan mode transitions by adding more aliases, enabling contextual exit confirmations, and providing user guidance. (02bf1476) (@vinhnx)
- Set custom terminal title for VT Code TUI (9459f366) (@vinhnx)
- Migrate changelog generation to git-cliff and update related documentation (34d07804) (@vinhnx)
- Add support for additional tree-sitter languages and make them optional (b1c01414) (@vinhnx)
- Make tree-sitter language parsers optional and update dependencies to disable default features (5e268705) (@vinhnx)
- Implement dynamic terminal title updates based on agent activity (8c29b28f) (@vinhnx)
- Add terminal notifications for agent events and syntax highlighting for bash commands in TUI (d54ba169) (@vinhnx)
- Add diagnostics visibility toggle in transcript for debug builds (2b339962) (@vinhnx)
- Include git version info in long_version and add tests (8a1fe108) (@vinhnx)
- Add snapshot files for history picker and slash widget tests (b397909d) (@vinhnx)
- Implement TUI log capture feature and enhance panic hook handling (89813122) (@vinhnx)
- Integrate color-eyre for enhanced error reporting in debug mode (05591670) (@vinhnx)
- Implement automatic context compaction and update related configurations (724af2c3) (@vinhnx)
- Enhance system prompt with detailed plan structure and error handling instructions (ec3fe254) (@vinhnx)
- Remove subagents and agent teams, update related configurations and documentation (81b135a9) (@vinhnx)
- Add Windows user profile root exclusions and enhance sensitive path matching (0a024067) (@vinhnx)
- Add new tools and policies for cargo and git commands (110d0326) (@vinhnx)
- Implement continuation arguments for tool execution and enhance read chunk handling (19db1b72) (@vinhnx)
- Enhance command execution tracking with arguments and output aggregation (80f4e11e) (@vinhnx)
- Refactor tool execution events to support new output and invocation structures (446a354c) (@vinhnx)
- Add tool output started event and integrate into streaming progress callback (d2551a50) (@vinhnx)
- Add support for OpenAI o3 and o4-mini models in Responses API (38f17b54) (@vinhnx)
- Enhance OpenAI Responses API support with hosted tools and validation (a17b42c8) (@vinhnx)
- Enhance shell execution approval process and protect writable metadata directories (74d40175) (@vinhnx)
- Implement thread management and session handling improvements (e775bb58) (@vinhnx)
- Enhance session management with loaded skills and metadata handling (cb99499b) (@vinhnx)
- Add checksum for vtcode-0.86.11-aarch64-apple-darwin.tar.gz (47a2c486) (@vinhnx)
- Implement feature gating for tools and refactor related components (5f530f6d) (@vinhnx)
- Refactor pattern detection logic, improve concurrency with `parking_lot`, and optimize performance with `SmallVec` and pre-allocated buffers. (28d0d534) (@vinhnx)
- Optimize tool execution tracking, file path parsing, and grep result handling, while preventing memory leaks in stats. (c319aebe) (@vinhnx)
- Add Miri check to the `check.sh` script and update documentation for detecting undefined behavior. (15f3d5c1) (@vinhnx)
- Add NVIDIA Nemotron 3 Super (free) model support (c3c203a0) (@vinhnx)
- Add live editor context and transcript file links (c330e4b3) (@vinhnx)
- Add Nemotron-3-Super model support in Ollama provider and constants (2602a7fc) (@vinhnx)
- Add context awareness support and related tests for Anthropic models (0ff5066e) (@vinhnx)
- Add context window size support to header context and related components (ba2faa13) (@vinhnx)
- Implement effective context window size handling for models (7ab580b4) (@vinhnx)
- Add OpenAI hosted shell support and validation checks (46031529) (@vinhnx)
- Enhance tool permission handling with new prompt options and decision logic (d8f448ba) (@vinhnx)
- Add OpenAI hosted shell configuration and validation for skills (c0383c1d) (@vinhnx)
- Implement context-generic provider wiring and enhance provider configuration (6c5d1910) (@vinhnx)
- Add prompt metadata to SnapshotMetadata and enhance hydration logic (79ca0b4b) (@vinhnx)
- Enhance OpenAI provider with max output tokens and previous response ID handling (f4e5237e) (@vinhnx)
- Enhance LLM request handling with non-streaming support and structured history integration (4180821a) (@vinhnx)
- Update default model to gpt-5.3-codex and enhance reasoning handling in OpenAI provider (6f8b0c13) (@vinhnx)
- Add manual input support for OpenAI OAuth flow and enhance session handling (ad2107c2) (@vinhnx)
- Enhance OpenAI Responses API handling and update configuration (5d5d55af) (@vinhnx)
- Enhance error handling by extracting human-readable messages from various JSON formats (b9d46cc2) (@vinhnx)
- Collapse large JSON outputs in TUI messages and adjust plan mode policies to allow. (80eed12d) (@vinhnx)
- Improve OAuth URL display by printing it on a separate line and remove related TODO item. (89d95a8a) (@vinhnx)
- Highlight file paths in markdown output (b50a3d62) (@vinhnx)
- Enhance OAuth flow with clickable URLs and improved user guidance (d5e9afa9) (@vinhnx)
- Implement transcript link click handling and update keyboard protocol flags (12e48c6d) (@vinhnx)
- Enhance mouse selection with double-click detection and clipboard functionality (25bae2b5) (@vinhnx)
- Implement explicit copy request for mouse selection and update Homebrew publishing process (21f2f1da) (@vinhnx)
- Update styling for shimmer effect and change UI theme to ayu (53ca8d5e) (@vinhnx)
- Add MiniMax-M2.7 model and update related configurations (3eeb3664) (@vinhnx)
- Add thread safety primitives including `SendWrapper` and enable optional concurrency for native plugins via a `thread_safe` flag. (86c0c612) (@vinhnx)
- Add configurable permissions mode (358f53fe) (@vinhnx)
- Add Claude-style compatibility for allowed and disallowed tools in permissions (bad5ba15) (@vinhnx)
- Improve tool output handling by ensuring summary headers are always recorded and refining stdout removal logic for specific tool types. (83e59742) (@vinhnx)
- Add hosted shell network policy support (def96551) (@vinhnx)
- Add support for OpenAI service tier "flex" and update related configurations (275f88c8) (@vinhnx)
- Add support for inline file inputs in OpenAI provider and enhance file handling (c07e0ac4) (@vinhnx)
- Add shared lightweight model routing (4e41e088) (@vinhnx)
- Implement preferred lightweight model variants and update configuration settings (2112d993) (@vinhnx)
- Implement collect_single_response function for streamlined LLM interactions and refactor related code (a863561c) (@vinhnx)
- Add interactive OpenAI manual compact controls (6660de5d) (@vinhnx)
- Implement transcript link detection and wrapping functionality (9cfd3e01) (@vinhnx)
- Implement link click throttling and improve mouse event handling (f6746ce3) (@vinhnx)
- Implement ast-grep command checks and improve initialization process (11e1d04f) (@vinhnx)
- Add Anthropic native memory and tool result clearing (7b7124f2) (@vinhnx)
- Update notification delivery mode to desktop (80b49f12) (@vinhnx)
- Combine native clearing with compaction (4edb1f65) (@vinhnx)
- Implement orchestrated harness mode with contract evaluation (b437d588) (@vinhnx)
- Enhance scorecard validation and add tests for missing/invalid criteria (29e3df0a) (@vinhnx)
- Update tool policies and add new tools with appropriate permissions (b7e51521) (@vinhnx)
- Enable ATIF trajectory export and update related configurations (47a230cc) (@vinhnx)
- Add mouse pointer shape handling and link/text selection detection (3f654089) (@vinhnx)
- Implement clipboard copy functionality and add copy notification (5a28a4be) (@vinhnx)
- Strip underlines from local file links in inline UI and add related tests (728eadc6) (@vinhnx)
- Enhance CLI progress reporting with terminal clear functionality (a600a3b4) (@vinhnx)
- Add Z.AI GLM-5.1 model support across various providers and update related tests (645df2e9) (@vinhnx)
- Enhance tool policies and workspace header signals for improved user experience (07910f3c) (@vinhnx)
- Add meta handling to elicitation response in LoggingClientHandler (0326ef68) (@vinhnx)
- Add checksum generation and validation to release process (6708fec0) (@vinhnx)
- Add generic action required status handling in overlays (6d7ffdad) (@vinhnx)
- Add thinking_display option for Anthropic provider to control API response format (0dd61648) (@vinhnx)
- Implement /effort command to set model reasoning effort level (11fbc317) (@vinhnx)
- Overhaul workspace check and test flow (bd4915c0) (@vinhnx)
- Add handling for reasoning signature in UI and ACP streams (9049a09a) (@vinhnx)
- Implement memories subsystem with configuration options (021aa664) (@vinhnx)
- Add OpenCode Zen and OpenCode Go providers and models (42bbea49) (@vinhnx)
- Add request_user_input_enabled flag to system prompt context and update related logic (ad39062d) (@vinhnx)
- Update Moonshot model references and configurations (365c522f) (@vinhnx)
- Update Moonshot model references to use DEFAULT_MODEL and add presets (794ee1a5) (@vinhnx)
- Add Kimi K2.6 model support in Ollama and Hugging Face providers (dbb69fc0) (@vinhnx)
- Add Kimi K2.6 support (d0a6a254) (@vinhnx)
- Add Kimi K2.6 to OpenRouter presets (04ca6808) (@vinhnx)
- Add MoonshotAI Kimi K2.6 model support (8882cf58) (@vinhnx)
- Add Ollama cloud provider support for DeepSeek V4 Flash (92908fe3) (@vinhnx)
- Add DeepSeek V4 Pro and Flash cloud models support (973ec7ff) (@vinhnx)
- Update tool policies and default model configuration (fea8947a) (@vinhnx)
- Enhance tool call safety validation and refactor permission checks (87f47eec) (@vinhnx)
- Add tool budget warning handling and input reset functionality (bea41f6c) (@vinhnx)
- Add tool call validation with session limit handling in CopilotRuntimeHost (a4d420b6) (@vinhnx)
- Add support for GPT-5.5 model and update related configurations (3296c4ce) (@vinhnx)
- Update default model and add ast-grep tool support (d2599e7a) (@vinhnx)
- Enhance local agents functionality and improve demo scripts (fea4eaf2) (@vinhnx)
- Implement caching for lightweight skill metadata and system prompt layers (4ba51c73) (@vinhnx)
- Enhance ownership management and borrowing practices across tools and registries (c606e2d4) (@vinhnx)
- Add MCP lifecycle controls and related tool functionalities (c186f725) (@vinhnx)
- Update tool policies and available tools in tool-policy.json (cd016ae5) (@vinhnx)
- Add tempfile dependency and enhance credential file permissions handling (6c198b3a) (@vinhnx)
- Enhance tool execution safety and prompt alignment handling (60c35814) (@vinhnx)
- Add DeepSeek V4 Flash and Pro models to Hugging Face integration (f7aac344) (@vinhnx)
- Implement auto-scroll toggle functionality and update related key bindings (b9e91ef5) (@vinhnx)
- Update model configs and wire models list through custom provider (76f5146c) (@vinhnx)
- Add 'file' alias for file path parameters across tools (381be8e5) (@vinhnx)
- Update tool policies and enhance DeepSeek provider handling (afaad984) (@vinhnx)
- Add new tools and update policies in tool-policy.json; refactor diff rendering to use standard ANSI colors (991d1590) (@vinhnx)
- Enhance tool output handling and add command summary recording (66d2691d) (@vinhnx)
- Add DSML parsing functionality and update reasoning effort configuration (fd749d7c) (@vinhnx)
- Enhance tool policy management and improve recovery handling (e4ef7275) (@vinhnx)
- Expose RecoveryPhase enum and improve recovery contract violation handling (30bde025) (@vinhnx)
- Implement DSML markup stripping and enhance navigation loop guidance (9e6b10a6) (@vinhnx)
- Remove tool policy JSON and enhance auto-approval logic in permission prompts (bfea75d4) (@vinhnx)
- Add hide header option for TUI to enhance user interface customization (b73d94fb) (@vinhnx)
- Improve tool policy permission management and remove auto-scroll functionality (0b833ca9) (@vinhnx)
- Enhance interim text continuation logic and improve conclusive marker detection (b76b7675) (@vinhnx)
- Refactor interim text continuation logic and enhance conclusive marker detection (2e5b96ea) (@vinhnx)
- Add command word splitting functionality for shell approval commands (3467ab3d) (@vinhnx)
- Optimize vector initializations and add #[must_use] annotations for clarity (e90ee59b) (@vinhnx)
- Update user message handling to use string slice for efficiency (854f732f) (@vinhnx)
- Enhance continuation evaluation with relaxed intent handling and add inline preview for spooled outputs (0d08141a) (@vinhnx)
- Add DeepSeek balance retrieval and display functionality (9c6aacc8) (@vinhnx)
- Enhance navigation loop detection and add DeepSeek balance tracking (5e9bdf89) (@vinhnx)
- Update balance tracking for multiple providers and refactor related logic (d3739bd3) (@vinhnx)
- Improve Linux installation logic and enhance balance fetching timeout handling (8a51b9a0) (@vinhnx)
- Update default model and add support for codex provider (1f361157) (@vinhnx)
- Add Mistral AI models and provider integration (81051c7a) (@vinhnx)
- Add Xiaomi MiMo models and provider integration (fa121dd7) (@vinhnx)
- Add HTTP headers for OpenRouter requests (66d456a9) (@vinhnx)
- Add support for mid-conversation system messages in Claude Opus 4.8 (076d8162) (@vinhnx)
- Add Laguna XS.2 model support and update configurations (5782eb5b) (@vinhnx)
- Add Poolside provider and models support (4b9aa40d) (@vinhnx)
- Add StepFun provider and model integration (c3f73b10) (@vinhnx)
- Enhance deserialization for optional strings and update PlannerResponse structure (3d6643d2) (@vinhnx)
- Add start and stop event stream commands to InlineCommand (2897d600) (@vinhnx)
- Add rebindable keybinding system with config integration (ddc2d952) (@vinhnx)
- Implement environment variable for non-interactive trust management (ed395835) (@vinhnx)
- Add workspace trust management details to documentation (a135f12b) (@vinhnx)
- Enable auto-trust for workspace in non-interactive mode (895437d8) (@vinhnx)
- Update default model and increase max tool calls per turn (7f6c3414) (@vinhnx)
- Add new models for DeepInfra and Novita inference providers (65527ea4) (@vinhnx)
- Add support for launching external editor with draft text (932f7197) (@vinhnx)
- Add Next.js agent evaluation details and leaderboard status (7f2faf53) (@vinhnx)
- Update README for improved clarity and structure (bb4ee130) (@vinhnx)
- Enhance badge section in README for better visibility and organization (30515213) (@vinhnx)
- Update README for improved clarity and organization (33c4a589) (@vinhnx)
- Add release packaging automation and shell completions support (575e4449) (@vinhnx)
- Add launch instruction after successful update installation (ce65af75) (@vinhnx)
- Implement Readline-style keybindings for input editing and navigation (b5572979) (@vinhnx)
- Enhance modal key handling and update quick help lines (08a09dad) (@vinhnx)
- Add MiniMax-M3 model with multimodal capabilities and update default model (64b27402) (@vinhnx)
- Add help modal for keyboard shortcuts and update related tests (20b360a1) (@vinhnx)
- Add MiniMax-M3 cloud model and update configurations (4d8b2cd2) (@vinhnx)
- Feat(paths): add expand_tilde function for home directory expansion (f92ab297) (@vinhnx)
- Add support for aarch64 target in Linux and update Homebrew formula (61f9f979) (@vinhnx)
- Add Xiaomi MiMo V2.5 and V2.5 Pro models with updated documentation and default settings (608b7fca) (@vinhnx)
- Add Evolink provider (OpenAI-compatible multi-model gateway) (bdc4c753) (@EvoLinkAI)
- Update file palette tree defaults and styling (3af4701b) (@vinhnx)
- Add Evolink provider support and update related documentation (e9577426) (@vinhnx)
- Update OpenAI function parameter sanitization and change default model in configuration (8871c397) (@vinhnx)
- Add new Evolink models and update configurations (8aafefb9) (@vinhnx)
- Add NVIDIA Nemotron 3 Ultra model and update provider logic (b27b6cbf) (@vinhnx)
- Add NVIDIA Nemotron 3 Ultra model and update related configurations (bff99dc1) (@vinhnx)
- Enhance JSON parsing to recover from truncated input (989d3bba) (@vinhnx)
- Add support for Nemotron 3 Ultra and Gemma 4 models with updated configurations (16e6026d) (@vinhnx)
- Add DRY CLI messages helper and detailed clap help for agents (f37e27c8) (@vinhnx)
- Enhance tool permission handling and improve ANSI escape sequence detection (b6e069ce) (@vinhnx)
- Enrich ast-grep structural search with metaVariables, byteOffset, and url fields (5dcbc31f) (@vinhnx)
- Implement subagent lifecycle tool policy and enhance subagent controller notifications (017ca5bf) (@vinhnx)
- Add support for new package managers and improve version handling (9dda3341) (@vinhnx)
- Implement strip_textual_tool_call_regions function and update related error messages (c62e4253) (@vinhnx)
- Enhance recovery handling for tool-call markers and improve documentation (e30b9e68) (@vinhnx)
- Add Claude Fable 5 with fallback credit, remove deprecated models (d4ed7872) (@vinhnx)
- Enhance Anthropic provider with new request handling and response parsing (73002a49) (@vinhnx)
- Add support for CLAUDE_OPUS_4_8 in thinking profile and update test case (24fc08e8) (@vinhnx)
- Add README for vtcode-ui crate with overview and usage instructions (b3bea69a) (@vinhnx)
- Add AGENTS.md for vtcode-ui with modules, rules, and gotchas (a2556300) (@vinhnx)
- Add vtcode-ui to publish sequence and update CLI documentation (bb7228ed) (@vinhnx)
- Add Kimi K2.7 Code model and update related configurations (5f74a144) (@vinhnx)
- Implement double-Esc detection for rewind functionality (ca7a521b) (@vinhnx)
- Add MiniMax-M3 model via Novita inference provider and update related configurations (ecdd32f4) (@vinhnx)
- Feat/refactor OpenAI chatgpt Rig (8a57c899) (@kernitus)
- Add compile-time validation for tool name constants and integrate ctor for pre-main checks (394677b2) (@vinhnx)
- Add Kimi K2.7 Code model support across configurations and descriptions (1767daec) (@vinhnx)
- Implement theme preview functionality and update default agent name (5b262e14)
- Add agent identity labels and apply them to system prompts (3ee26a3d)
- Update chat input placeholders and footer hints to include agent switching (4af057cb)
- Add primary agent color field to header context and update related tests (15354220)
### Other Changes
#### CI
- Allow linux packaging fallback without ghostty runtime (da97cd00) (@vinhnx)
- Fix x86_64-unknown-linux-gnu aws-lc-sys gcc-9 panic (58a30561) (@vinhnx)
- Enable Windows builds by default in release workflows (7bce5b8d) (@vinhnx)
- Improve Rust toolchain management with MSRV check and nightly smoke test (9909ad4b) (@vinhnx)
#### Other
- Fix provider inference for router (ec7a38a4) (@vinhnx)
- Updates TODO (eeecc852) (@vinhnx)
- Refactor Gemini provider to use shared abstractions (824e49a0) (@vinhnx)
- Fix Gemini reasoning config placement (3f182136) (@vinhnx)
- Handle provider failures without trapping the runloop (270449c5) (@vinhnx)
- Updates (10677805) (@vinhnx)
- Updates (585403bd) (@vinhnx)
- Updates (fe9987c1) (@vinhnx)
- Wrap Gemini tool responses with content parts (1c34c6b6) (@vinhnx)
- Updates (d8c77da8) (@vinhnx)
- Updates (2b74c288) (@vinhnx)
- Add streaming token support via async-stream (61623d3e) (@vinhnx)
- Gate ask streaming to supported providers (d54c15ab) (@vinhnx)
- Handle static ask requests without streaming (5dfd3392) (@vinhnx)
- Gate streaming requests by provider capability (eb9ce789) (@vinhnx)
- Fix OpenRouter streaming delta handling (8591b2e2) (@vinhnx)
- Improve streaming UX and reasoning aggregation (75b4118d) (@vinhnx)
- Fix streaming parsing and reasoning formatting (0957b690) (@vinhnx)
- Fix Gemini streaming event processing (33ef87bc) (@vinhnx)
- Improve Gemini streaming event parsing (41c8a86e) (@vinhnx)
- Handle partial Gemini streaming events (a16e65cb) (@vinhnx)
- Fix OpenRouter SSE delimiter parsing (7ba24019) (@vinhnx)
- Improve streaming parsing and refine prompt guardrails (d58c9574) (@vinhnx)
- Fix Gemini streaming incremental tokens (af5c52a3) (@vinhnx)
- Stream reasoning tokens from OpenRouter (6fe80a2c) (@vinhnx)
- Updates (30a5e1cc) (@vinhnx)
- Update config (5d94de5c) (@vinhnx)
- Expose reasoning traces in chat transcript (b0ecf226) (@vinhnx)
- Ensure reasoning fallback logs to transcript (dfe51286) (@vinhnx)
- Fix OpenRouter reasoning trace extraction (6b104a85) (@vinhnx)
- Disable reasoning trace handling (675b5a85) (@vinhnx)
- Restore reasoning effort configuration usage (3a931dd1) (@vinhnx)
- Add configurable reasoning effort handling (e15574e6) (@vinhnx)
- Refactor session banner rendering for improved clarity and consistency (118cb0e3) (@vinhnx)
- Add workspace trust prompt before chat sessions (c1554988) (@vinhnx)
- Refine workspace trust prompt and banner (24bb2e07) (@vinhnx)
- Handle EOF in workspace trust prompt (c34533a7) (@vinhnx)
- Updates (db199a49) (@vinhnx)
- Updates (64651f65) (@vinhnx)
- Updates (1a0458c2) (@vinhnx)
- Add cfonts dependency, update README image, and enhance UI rendering (2237be84) (@vinhnx)
- Updates (a773b64f) (@vinhnx)
- Improve system prompt context discipline (7d1bee3b) (@vinhnx)
- Fix xAI provider integration (32b3417f) (@vinhnx)
- Add markdown rendering with syntax highlighting for agent output (b1b3fafa) (@vinhnx)
- Restore incremental response rendering with markdown updates (ca19c737) (@vinhnx)
- Improve curl tool output and Ctrl+C summary handling (d30ab220) (@vinhnx)
- Add curl policy defaults and show HITL banner state (20598a96) (@vinhnx)
- Enhance file search with git-aware fuzzy matching (eb33d5a5) (@vinhnx)
- Ensure file size limit applies to symlinks (d76ae06e) (@vinhnx)
- Updates (445d9c1c) (@vinhnx)
- Updates (cc3177f2) (@vinhnx)
- Updates (04c0a868) (@vinhnx)
- Updates (c266ed76) (@vinhnx)
- Updates (625343cb) (@vinhnx)
- Improve curl tool output and Ctrl+C summary handling (cdc1649d) (@vinhnx)
- Add curl policy defaults and show HITL banner state (65dad4dc) (@vinhnx)
- Enhance file search with git-aware fuzzy matching (32027a08) (@vinhnx)
- Ensure file size limit applies to symlinks (9c329b68) (@vinhnx)
- Updates (92718de7) (@vinhnx)
- Updates (2b57c3fb) (@vinhnx)
- Updates (584a23af) (@vinhnx)
- Updates (69f6119f) (@vinhnx)
- Refine onboarding guideline rendering (5c76bd59) (@vinhnx)
- Refine plan rendering and tracking (21f1d36b) (@vinhnx)
- Fix HITL prompt rendering in iocraft UI (cc870b45) (@vinhnx)
- Update TODO.md (0eabfa71) (@vinhnx)
- Update TODO.md (32af312e) (@vinhnx)
- Revise VT Code description and API key setup (6f6247d8) (@vinhnx)
- Refine ratatui chat layout (403ef80b) (@vinhnx)
- Integrate scrollview into ratatui transcript (aaddc846) (@vinhnx)
- Rework ratatui transcript scrolling (b1e90df9) (@vinhnx)
- Refine ratatui message blocks and scrolling (1142ca8a) (@vinhnx)
- Refactor ratatui transcript rendering to match example layout (be494646) (@vinhnx)
- Refine ratatui scrolling and remove transcript echoes (9b5a979e) (@vinhnx)
- Add mouse wheel scrolling and bottom status bar (f7c1bae0) (@vinhnx)
- Add slash suggestions and PTY panel to ratatui chat UI (fd3d94b4) (@vinhnx)
- Refine ratatui chat layout and status metadata (66e1b645) (@vinhnx)
- Update config (f815d849) (@vinhnx)
- Update TODO.md with new tasks and resources (fa4a701e) (@vinhnx)
- Updates (7175a84e) (@vinhnx)
- Updates (4fdef029) (@vinhnx)
- Updates (43cc3211) (@vinhnx)
- Refine ratatui transcript interactions (170fe256) (@vinhnx)
- Revamp ratatui transcript layout (de2f7a6b) (@vinhnx)
- Tune ratatui welcome and tool styling (78ad06e0) (@vinhnx)
- Updates (2f5f6358) (@vinhnx)
- Updates (7952021e) (@vinhnx)
- Updates (d568f542) (@vinhnx)
- Fix prefix ❯ (436cd902) (@vinhnx)
- Fix npm publishing path and validation (29482be0) (@vinhnx)
- Remove API key inference details from README (02bdd062) (@vinhnx)
- Update TODO.md with reference link (551fbae6) (@vinhnx)
- Improve ratatui surface handling and PTY widget (d9508743) (@vinhnx)
- Improve tool call display and add conversation history scrolling (c6fee0bd) (@vinhnx)
- Allow scrolling across entire transcript history (aa426e53) (@vinhnx)
- Document modular TUI refactor (ec4e4610) (@vinhnx)
- Handle inline surface without TTY requirements (a5dcf623) (@vinhnx)
- Updates (49502e5f) (@vinhnx)
- Bump npm package version to 0.13.7 (f2b68000) (@vinhnx)
- Update Homebrew formula to version 0.13.7 (21ab51c5) (@vinhnx)
- Add automated binary build and upload script (e06539db) (@vinhnx)
- Update Homebrew formula to version 0.13.8 (064a9d17) (@vinhnx)
- Updates (b8628c21) (@vinhnx)
- Updates (f6b1db1d) (@vinhnx)
- Update TODO.md (b3018e86) (@vinhnx)
- Update TODO.md with new tasks and document links (875edf4a) (@vinhnx)
- Update README with new screenshot and enhance UI feature descriptions (18f5489d) (@vinhnx)
- Add Ratatui badge to README for enhanced visibility (006684e0) (@vinhnx)
- Update dependencies in Cargo.toml for vtcode and vtcode-core (8bbe68b4) (@vinhnx)
- Updates (baa67855) (@vinhnx)
- Updates (fcb4fab2) (@vinhnx)
- Fix OpenAI tool payload and restore mouse capture (47c1e5c1) (@vinhnx)
- Fix OpenRouter responses parsing for GPT-5 Codex (eb57ddf6) (@vinhnx)
- Update README with new screenshot and enhance UI feature descriptions (23b04bf9) (@vinhnx)
- Updates (46df718c) (@vinhnx)
- Fix OpenRouter responses parsing for GPT-5 Codex (1e49f09c) (@vinhnx)
- Remove width attribute from demo image (bf1f1a6e) (@vinhnx)
- Updates (3501a578) (@vinhnx)
- Revise VT Code description in README (41914c9e) (@vinhnx)
- Updates TODO (15b26a56) (@vinhnx)
- Integrate Catppuccin palette into theme registry (516f0174) (@vinhnx)
- Refine Catppuccin registration (d83ccaa3) (@vinhnx)
- Updates (9ddbab43) (@vinhnx)
- Add session archiving and listing commands (3cdd2eb9) (@vinhnx)
- Enhance session archive summaries with message previews (75cad42b) (@vinhnx)
- Add labeled chat panels and left-aligned agent messages (964ff9cf) (@vinhnx)
- Enhance terminal experience description in README (f46d67b9) (@vinhnx)
- Enhance README with updated links and descriptions (828e427b) (@vinhnx)
- Enhance VT Code description and link formatting (e90dbd07) (@vinhnx)
- Add links to vtcode Crate and Documentation (bfebcaad) (@vinhnx)
- Remove links section from README (e020060f) (@vinhnx)
- Update TODO.md with VS Code extension link (a7a392f8) (@vinhnx)
- Update TODO.md with homebrew issue details (2ab8fac2) (@vinhnx)
- Implement prompt caching across multiple LLM providers (e9701859) (@vinhnx)
- Add comprehensive prompt caching support across multiple LLM providers (98867e0a) (@vinhnx)
- Update README.md (9bca5beb) (@vinhnx)
- Update README.md (644d46fa) (@vinhnx)
- Update README.md (c1bb5f21) (@vinhnx)
- Update README.md (e338d929) (@vinhnx)
- Update TODO.md with new reference and tasks (d228e917) (@vinhnx)
- Update README (9b25b969) (@vinhnx)
- Updates (9a2719bf) (@vinhnx)
- Updates (5895a9b3) (@vinhnx)
- Updates (38db6a8a) (@vinhnx)
- Updates (b9c59f98) (@vinhnx)
- Swap ratatui session to termion backend (508c320d) (@vinhnx)
- Refactor inline TUI to tuirealm architecture (58275ecf) (@vinhnx)
- Refine inline tui transcript layout (8a9d664d) (@vinhnx)
- Gracefully cancel MCP providers on shutdown (eaddff5f) (@vinhnx)
- Refine inline tui layout and tool panel output (37f0627f) (@vinhnx)
- Limit inline TUI viewport height via config (41b75d42) (@vinhnx)
- Simplify MCP tool output formatting (1f86142f) (@vinhnx)
- Refine inline transcript formatting (eae880f0) (@vinhnx)
- Simplify inline TUI layout (dba7b607) (@vinhnx)
- Refactor inline TUI to component system (1c8fc4ef) (@vinhnx)
- Handle streaming lines in inline TUI (d8f2ecb2) (@vinhnx)
- Simplify inline ratatui session layout (163ccdea) (@vinhnx)
- Refine inline UI naming after dropping Ratatui (b0091360) (@vinhnx)
- Simplify inline prefixes for agent transcript (9eb174fe) (@vinhnx)
- Fix termion session cleanup (3eedd3cc) (@vinhnx)
- Add missing MCP event helpers for tests (6c1b04b6) (@vinhnx)
- Update TODO.md (e83af2dd) (@vinhnx)
- Revise README for VT Code features and installation (d79e8071) (@vinhnx)
- Change default provider to OpenAI and update model (8eb08c97) (@vinhnx)
- Fix formatting for OpenAI caching details in README (7d8a0909) (@vinhnx)
- Simplify VT Code description in README (2043c94b) (@vinhnx)
- Update TODO list with new submission tasks (5a81a2a9) (@vinhnx)
- Update TODO.md with new resource links (305ec05a) (@vinhnx)
- Update funding sources in FUNDING.yml (0d610ed3) (@vinhnx)
- Update TODO.md with TUI tasks and improvements (e6fd113f) (@vinhnx)
- Update TODO.md to remove outdated Homebrew info (6b3ba4a2) (@vinhnx)
- Updates (ca0a6c00) (@vinhnx)
- Updates (e42245ee) (@vinhnx)
- Updates (ff439027) (@vinhnx)
- Update tool policies and UI configurations for improved functionality (840484e1) (@vinhnx)
- Enhance MCP integration and UI features (101d47e2) (@vinhnx)
- Fix MCP allowlist overrides and document integration (ef10aa56) (@vinhnx)
- Improve MCP client handshake and curl summary (3faa568d) (@vinhnx)
- Restore MCP logging handler for rmcp client (84ddcb16) (@vinhnx)
- Fix MCP meta payload handling (e30f35b6) (@vinhnx)
- Improve MCP tool error reporting (2214a625) (@vinhnx)
- Fix inline scrollback truncation (caa0ec9e) (@vinhnx)
- Maintain inline scroll position when streaming (f82daa98) (@vinhnx)
- Clamp inline session scrollback to buffer bounds (526bdaf7) (@vinhnx)
- Refactor inline TUI to use ratatui (176972f4) (@vinhnx)
- Enhance inline TUI viewport and alternate screen (74278d77) (@vinhnx)
- Improve inline TUI scrollback handling (97e9a840) (@vinhnx)
- Improve ratatui scroll metrics handling (9d18132d) (@vinhnx)
- Refactor inline TUI to crossterm backend (fa66bdf3) (@vinhnx)
- Handle MCP tool aliases when hyphens are stripped (5d8dbdf9) (@vinhnx)
- Stabilize banner logo and unblock inline redraw (e17b7d2d) (@vinhnx)
- Remove inline TUI background color (ae429c96) (@vinhnx)
- Simplify session banner text (0f6c54cc) (@vinhnx)
- Improve welcome spacing and animate thinking spinner (374c826b) (@vinhnx)
- Updates (28b97bb5) (@vinhnx)
- Fix grammar and clarity in VT Code description (c15edf36) (@vinhnx)
- Update TODO.md (d3dbd780) (@vinhnx)
- Fix typo in README.md heading (b7859558) (@vinhnx)
- Update link for vtcode.toml in README (5110057f) (@vinhnx)
- Revise README for VT Code features and installation (17fabfea) (@vinhnx)
- Fix HTML formatting in README.md (cc7beddc) (@vinhnx)
- Add README about context engineering foundation of vtcode (2b8816af) (@vinhnx)
- Change default provider to OpenAI and update model (f1c76117) (@vinhnx)
- Fix formatting for OpenAI caching details in README (2e2de80c) (@vinhnx)
- Update README to remove Serena MCP mentions (5aeaa113) (@vinhnx)
- Simplify VT Code description in README (e6b44e26) (@vinhnx)
- Update TODO list with new submission tasks (0f434694) (@vinhnx)
- Update TODO.md with new resource links (8e8dcf30) (@vinhnx)
- Update funding sources in FUNDING.yml (7fe2e8da) (@vinhnx)
- Update TODO.md with TUI tasks and improvements (7b5ce742) (@vinhnx)
- Update TODO.md to remove outdated Homebrew info (8adc7c03) (@vinhnx)
- Update README (7441988f) (@vinhnx)
- Updates (a9bc8013) (@vinhnx)
- Updates (03dcb5b3) (@vinhnx)
- Updates (7c1560f4) (@vinhnx)
- Updates (6a480c6a) (@vinhnx)
- Update TODO.md with new implementation tasks (b4a8e6d8) (@vinhnx)
- Update README with vtcode.toml setup instructions (36bb5006) (@vinhnx)
- Update README with notes on configuration options (287349c1) (@vinhnx)
- Update TODO list and file organization (040c01ed) (@vinhnx)
- Adjust OpenAI reasoning model list (cc9c5100) (@vinhnx)
- Route OpenAI o-series models through Responses API (d438afd4) (@vinhnx)
- Add slash commands for init and config (d351dc77) (@vinhnx)
- Handle binary files during workspace indexing (36e632d6) (@vinhnx)
- Align OpenAI tool serialization with function wrapper (7f40294f) (@vinhnx)
- Improve inline TUI viewport responsiveness (732c65ec) (@vinhnx)
- Respect CLI model overrides (bd515d47) (@vinhnx)
- Use list widget for chat transcript (136613ec) (@vinhnx)
- Preserve newline wrapping in transcript list (86fa2653) (@vinhnx)
- Add word-wise cursor movement in inline chat input (dd50ae5c) (@vinhnx)
- Improve inline chat navigation modifiers (67e7c195) (@vinhnx)
- Updates (be91ff64) (@vinhnx)
- Prefill slash command selection into input (2350651d) (@vinhnx)
- Add slash command suggestion list with keyboard navigation (9e2ed36f) (@vinhnx)
- Updates release script (d49bd604) (@vinhnx)
- Updates (ebcd4e7f) (@vinhnx)
- Account for slash suggestion borders in layout (0bca709d) (@vinhnx)
- Align OpenAI tool serialization with function wrapper (f8e9fda5) (@vinhnx)
- Improve inline TUI viewport responsiveness (0894c9ef) (@vinhnx)
- Respect CLI model overrides (08ea7334) (@vinhnx)
- Use list widget for chat transcript (afaf3bd2) (@vinhnx)
- Preserve newline wrapping in transcript list (cd012840) (@vinhnx)
- Add word-wise cursor movement in inline chat input (ee6bf456) (@vinhnx)
- Improve inline chat navigation modifiers (e5b7a585) (@vinhnx)
- Updates (96e08729) (@vinhnx)
- Prefill slash command selection into input (8ca758f0) (@vinhnx)
- Add slash command suggestion list with keyboard navigation (6310077f) (@vinhnx)
- Updates release script (afac308a) (@vinhnx)
- Updates (ddd041c4) (@vinhnx)
- Account for slash suggestion borders in layout (a9c493b8) (@vinhnx)
- Handle Ctrl+C interrupts with cancel vs exit (6062b686) (@vinhnx)
- Preserve double Ctrl+C exit after cancellation (649508dd) (@vinhnx)
- Add background section to README (192ee87e) (@vinhnx)
- Update links and references in README.md (a3acef1b) (@vinhnx)
- Allow update_plan tool (3ee39f54) (@vinhnx)
- Update brew hash (9b2683af) (@vinhnx)
- Improve io (d65edd07) (@vinhnx)
- Fix release (70e0b7cb) (@vinhnx)
- Update Cargo.lock for version 0.15.4 (cd69203f) (@vinhnx)
- Upd (5d461d7c) (@vinhnx)
- Updates (f9cf48d8) (@vinhnx)
- Update release script (bc67f827) (@vinhnx)
- Add inline modal for human approval prompts (754621b6) (@vinhnx)
- Improve HITL modal dismissal and input sync (4ef9f69a) (@vinhnx)
- Remove HITL approval modal (70537bf9) (@vinhnx)
- Update TODO.md with important link and cleanup (74f4c49d) (@vinhnx)
- Updates TODO (9f0c1816) (@vinhnx)
- Update TODO.md with new context engineering links (67a9f1ef) (@vinhnx)
- Enhance context engineering (9db2638b) (@vinhnx)
- Add review of context engineering commit 9db2638 (584b28f6) (@vinhnx)
- Refine context engineering review action items (238d87f1) (@vinhnx)
- Integrate context curator and token budget into run loop (50f562cb) (@vinhnx)
- Stabilize tool policy availability ordering (a4ab59dc) (@vinhnx)
- Revamp welcome banner layout (396f8483) (@vinhnx)
- Add indicatif spinner to thinking status (91e7f7a5) (@vinhnx)
- Show VT Code version in welcome logo (fa7af354) (@vinhnx)
- Updates (6656a349) (@vinhnx)
- Fix OpenAI max completion tokens handling (785a02b9) (@vinhnx)
- Adding deepseek provider (9be6e728) (@oiwn)
- Adjust inline TUI header and layout (d14d4d44) (@vinhnx)
- Align inline header separators and rounded borders (16c0c844) (@vinhnx)
- Add inline session scrollbar and dynamic title layout (8466e00b) (@vinhnx)
- Updates tool policy (48a016a2) (@vinhnx)
- Refactor README.md for improved structure and clarity (0685f510) (@vinhnx)
- Update TODO with context engineering tasks (95a3e848) (@vinhnx)
- Gate onboarding placeholder on enabled flag (78e6759a) (@vinhnx)
- Treat onboarding chat placeholder as optional (8b9c54a6) (@vinhnx)
- Add toggle for TODO planning mode and style placeholder (0bb49198) (@vinhnx)
- Remove onboarding placeholder and add chat dividers (672d431b) (@vinhnx)
- Add agent quote prefix styling (f1417357) (@vinhnx)
- Enhance technical motivation section in README (667e4074) (@vinhnx)
- Add timeline toggle and refresh inline tool styling (0aaf787d) (@vinhnx)
- Adjust inline pane styling and remove agent star (346915a5) (@vinhnx)
- Expand default tool policies and welcome sections (92234c17) (@vinhnx)
- Simplify inline welcome banner controls (8bc20dc3) (@vinhnx)
- Update inline header layout and agent styling (01a248b5) (@vinhnx)
- Increase inline header height for two-line layout (02abb485) (@vinhnx)
- Remove transcript scrollbar rendering (431efa95) (@vinhnx)
- Make inline header height dynamic (8f6d3c59) (@vinhnx)
- Align theme default with config constant (3cf3a530) (@vinhnx)
- Improve sequential tool output styling (7f6f01db) (@vinhnx)
- Use renderer styles for sequential thinking headers (b14a4dde) (@vinhnx)
- Handle OpenRouter tool fallback (6473645b) (@vinhnx)
- Add xAI provider inference for CLI api key env (38a6cc47) (@vinhnx)
- Revise TODO.md with design links and styling tasks (be3e370c) (@vinhnx)
- Restore scroll view for transcript pane (825aaa18) (@vinhnx)
- Adjust transcript scroll padding (c88214e8) (@vinhnx)
- Align transcript scroll area with padding (ecf867a9) (@vinhnx)
- Avoid rendering offscreen transcript rows (56c2ed71) (@vinhnx)
- Optimize transcript rendering cache (e7ba38e3) (@vinhnx)
- Improve transcript reflow caching for streaming (64e0b9ca) (@vinhnx)
- Refine transcript scrolling and padding (18134655) (@vinhnx)
- Add optional Swift tree-sitter integration (919d400d) (@vinhnx)
- Propagate Swift unsupported errors (b056b16f) (@vinhnx)
- Stop Swift fallback when feature disabled (cbcd8627) (@vinhnx)
- Refine Zed ACP tool negotiation (2ff6236e) (@vinhnx)
- Document Zed ACP setup across guides and crate docs (96d56eae) (@vinhnx)
- Enable ACP and ZED configurations in vtcode.toml (7bcae249) (@vinhnx)
- Update dependencies and refine tree-sitter integration (92d9f440) (@vinhnx)
- Update agent provider and default model to use Claude Sonnet (97e30066) (@vinhnx)
- Add optional Swift tree-sitter integration (6aa11807) (@vinhnx)
- Propagate unsupported language detection errors (7aa8a40c) (@vinhnx)
- Ensure cargo-release dry runs stay non-executing (ddc6949e) (@vinhnx)
- Sanitize Gemini tool schemas (f47e8618) (@vinhnx)
- Improve MCP client logging and stderr handling (05cfcdc6) (@vinhnx)
- Refactor language initialization to use an immutable vector (dc6b7fe9) (@vinhnx)
- Add permission gating to ACP read file tool (61beb266) (@vinhnx)
- Refine ACP handshake capability handling (40347be1) (@vinhnx)
- Improve ACP cancellation handling (c6920aad) (@vinhnx)
- Harden ACP read_file validation (69fb3a0a) (@vinhnx)
- Add ACP plan tracking to Zed integration (be6bc693) (@vinhnx)
- Document ACP guarantees for Zed integration (8668f3d8) (@vinhnx)
- Clarify ACP documentation links (bb7dc2de) (@vinhnx)
- Fail closed when Zed permission requests error (731aede2) (@vinhnx)
- Fix anstyle-migration-tool Cargo.toml missing fields for cargo-release (f7835b15) (@vinhnx)
- Enhance release script with pre-release versioning options (b7355750) (@vinhnx)
- Update run_prerelease function to use correct cargo-release syntax (ebf05b19) (@vinhnx)
- Fix release.toml pre-release-replacements to target specific package (e6bd99c8) (@vinhnx)
- Revert release.toml to original after testing incorrect package field (54b43f35) (@vinhnx)
- Temporarily remove npm package.json replacement to test new script functionality (81fae09f) (@vinhnx)
- Fix cargo-release flag from --skip-publish to --no-publish (2e547a4c) (@vinhnx)
- Add changelogithub config and update release.toml for changelog generation (5518527b) (@vinhnx)
- Remove npm replacement temporarily to ensure changelog generation setup works (fd8d8736) (@vinhnx)
- Document changelog generation process in release script (fb5b4103) (@vinhnx)
- Add release process documentation with changelog generation info (519d8650) (@vinhnx)
- Remove emojis from changelogithub configuration (d34f9240) (@vinhnx)
- Update documentation to reflect emoji-free changelog titles (39a6e1be) (@vinhnx)
- Fix pre-release versioning to avoid duplicate alpha markers (3cbc3436) (@vinhnx)
- Fix formatting issues in README.md (d972142d) (@vinhnx)
- Revise VT Code Agent registration instructions (f12e4dc0) (@vinhnx)
- Remove cargo installation example for vtcode (b5c4dee1) (@vinhnx)
- Add custom model usage instructions for VT Code (16e77557) (@vinhnx)
- Clarify vtcode.toml review instructions (8bd2f6ad) (@vinhnx)
- Enhance release script with npm package version update functionality (426ffd67) (@vinhnx)
- Fix npm package version update logic and parameter passing (65d2436e) (@vinhnx)
- Add Zed agent client protocol badge to README (260ea66a) (@vinhnx)
- Update README with installation instructions (58ba0469) (@vinhnx)
- Update Zed ACP integration documentation (72743412) (@vinhnx)
- Add autonomous benchmark runner (0b7d2ca9) (@vinhnx)
- Add autonomous CLI execution mode (4db83468) (@vinhnx)
- Align benchmark runner with workspace tool config (396567f5) (@vinhnx)
- Normalize reasoning effort handling across providers (08274d2c) (@vinhnx)
- Refine inline model picker UI (6a178d22) (@vinhnx)
- Render slash command palette as overlay (ff042774) (@vinhnx)
- Fix API key env inference for non-default providers (1f1ac9a2) (@vinhnx)
- Handle default API key env override (f8dd8f16) (@vinhnx)
- Add npm package version update functionality with proper version calculation (fffc72bf) (@vinhnx)
- Update binary distribution messaging and finalize npm package version update (bfcfb6ea) (@vinhnx)
- Updates (df274de1) (@vinhnx)
- Remove git add from npm update function to avoid staged changes issue (2f8e19ec) (@vinhnx)
- Remove npm package replacement to avoid workspace processing issues (a34d6e5d) (@vinhnx)
- Add function to commit npm package update after cargo-release completes (ce6511d1) (@vinhnx)
- Change srgn permission from prompt to allow in vtcode.toml (bd0779a8) (@vinhnx)
- Add ACP list_files bridge for Zed (260ab2ad) (@vinhnx)
- Ensure ACP bridge enforces full workspace trust (9b52fc24) (@vinhnx)
- Apply vtcode tool policies in ACP bridge (57220a0f) (@vinhnx)
- Fix ACP shell execution and workspace path handling (f17d8600) (@vinhnx)
- Remove anstyle-migration-tool and related files from the project (34d3445e) (@vinhnx)
- Improve npm package version handling during release process (197d0cd6) (@vinhnx)
- Clean up TODO.md by removing obsolete entries (fa0d4f2d) (@vinhnx)
- Add npm package version update functionality with proper version calculation (537ad55d) (@vinhnx)
- Update binary distribution messaging and finalize npm package version update (feb748b1) (@vinhnx)
- Updates (5cb66608) (@vinhnx)
- Remove git add from npm update function to avoid staged changes issue (288b1459) (@vinhnx)
- Remove npm package replacement to avoid workspace processing issues (8352d225) (@vinhnx)
- Add function to commit npm package update after cargo-release completes (db53d6ce) (@vinhnx)
- Change srgn permission from prompt to allow in vtcode.toml (ecccb8c2) (@vinhnx)
- Add ACP list_files bridge for Zed (2b5bfd6a) (@vinhnx)
- Ensure ACP bridge enforces full workspace trust (d8dc4ed1) (@vinhnx)
- Apply vtcode tool policies in ACP bridge (94ffdfbf) (@vinhnx)
- Fix ACP shell execution and workspace path handling (86e81184) (@vinhnx)
- Remove anstyle-migration-tool and related files from the project (936997c8) (@vinhnx)
- Improve npm package version handling during release process (e0adcb16) (@vinhnx)
- Clean up TODO.md by removing obsolete entries (557249b5) (@vinhnx)
- Add ci/coverage workflows (47c747dd) (@oiwn)
- Add files for deny and typos (55684a4d) (@oiwn)
- Proper deny config (fe5b29ce) (@oiwn)
- Cargo deny should pass (a882e2b8) (@oiwn)
- Disable lint-typos and clippy in CI configuration (39087607) (@vinhnx)
- Prevent justifying lines inside fenced code blocks (eb689b12) (@vinhnx)
- Allow string plan steps and default statuses (7a69bbf9) (@vinhnx)
- Sanitize OpenRouter requests for tool-ineligible models (46d9e4e1) (@vinhnx)
- Fix OpenRouter tool filters for OpenAI minis (b6cf934b) (@vinhnx)
- Refresh OpenRouter model catalog (92c227da) (@vinhnx)
- Make OpenRouter model match arms exhaustive (5e156c5c) (@vinhnx)
- Add OpenRouter GLM 4.5 Air free constant (ffd0c8e2) (@vinhnx)
- Simplify welcome overview and slash command hints (f0762542) (@vinhnx)
- Revise contribution guidelines for pull requests (156546a0) (@vinhnx)
- Add Z.AI provider integration (ba3a4769) (@vinhnx)
- Enhance VT Code installation instructions (a1e70456) (@vinhnx)
- Add coverage for new models in picker (083a5c45) (@vinhnx)
- Align xAI reasoning detection with Grok 4 identifiers (aaf1c84a) (@vinhnx)
- Prioritize OpenAI and Anthropic in model picker (4237a104) (@vinhnx)
- Add workspace BYOK flow to model picker (c73581b2) (@vinhnx)
- Integrate secure BYOK prompt infrastructure (dae522f3) (@vinhnx)
- Keep input enabled for secure modal prompts (f092813f) (@vinhnx)
- Add DeepWiki badge to README (bd9b995a) (@vinhnx)
- Fix secure key modal input and improve model grouping (46871e62) (@vinhnx)
- Add search filtering to model picker modal (dbe455ea) (@vinhnx)
- Add usage tips and recommended actions highlights to onboarding (2eccb6d7) (@vinhnx)
- Improve TUI markdown rendering (d6975d0e) (@vinhnx)
- Improve TUI line clipping (ce66d9c0) (@vinhnx)
- Fix CRLF handling in wrap_line (598d58a5) (@vinhnx)
- Integrate portable PTY manager and tests (404648d1) (@vinhnx)
- Guard wait status match for non-Linux builds (1aa85d7b) (@vinhnx)
- Guard PTY command runner on non-Unix targets (060b3fb1) (@vinhnx)
- Update TODO.md (c668ae47) (@vinhnx)
- Update README.md (1bd1499b) (@vinhnx)
- Add QR buymeacoffee (9b1f642f) (@vinhnx)
- Add support section for VT Code development (d9f20eec) (@vinhnx)
- Condense slash command welcome hint (9074c080) (@vinhnx)
- Add Clippy lints for stricter code quality (63385eb3) (@vinhnx)
- Updates config (ee5d4e45) (@vinhnx)
- Enhance ACP tool call updates (e4a4af1b) (@vinhnx)
- Replace rexpect PTY helper with portable_pty implementation (d53c3215) (@vinhnx)
- Add GitHub Packages publishing instructions and scripts (7e49fd2a) (@vinhnx)
- Show git branch status in chat input footer (963da007) (@vinhnx)
- Adjust inline input status layout (0148859f) (@vinhnx)
- Throttle inline git status refresh (646bacdb) (@vinhnx)
- Handle missing tokenizers gracefully (f66f9676) (@vinhnx)
- Record estimated tokens when tokenizer fallback occurs (47dc9824) (@vinhnx)
- Handle MCP pagination and protocol validation (f06711ca) (@vinhnx)
- Propagate MCP initialize params to transport handshake (9759d541) (@vinhnx)
- Add exec mode with structured events (79c43bd8) (@vinhnx)
- Stream exec events live (4bae0a43) (@vinhnx)
- Add Ollama provider to LLM client and registry (1080db03) (@vinhnx)
- Extract CLI startup context builder (a58b705c) (@vinhnx)
- Refine router classification and provider inference (2a27a309) (@vinhnx)
- Refactor context curator token estimation (80f11d17) (@vinhnx)
- Refactor tool registry inventory and policy layers (9e87b14e) (@vinhnx)
- Refactor LLM factory registration helpers (b9c6bf81) (@vinhnx)
- Centralize configuration defaults and validation (c13aa413) (@vinhnx)
- Add system benchmarks and stabilize startup tests (a8bb9e5f) (@vinhnx)
- Add plan for extracting reusable crates (0a29c267) (@vinhnx)
- Fix docs warnings for docs.rs build (76687e3c) (@vinhnx)
- Improve build docs.rs (c730383e) (@vinhnx)
- Fix Ollama base URL (291ffa78) (@vinhnx)
- Add support for custom Ollama model and update configuration (d4a7a248) (@vinhnx)
- Add Ollama support and update documentation for local model integration (d360e6e3) (@vinhnx)
- Update default LLM provider config (dc8deac8) (@vinhnx)
- Vendor mcp-types for docs.rs (ede286bf) (@vinhnx)
- Restore MCP_SCHEMA_VERSION alias (f61279d5) (@vinhnx)
- Improve Zed ACP list_files tool path handling (760ab4cc) (@vinhnx)
- Update badge links in README.md (7eb9f21d) (@vinhnx)
- Remove duplicate badges from README (49513893) (@vinhnx)
- Fix OpenAI tool serialization with top-level name (eac48ec5) (@vinhnx)
- Improve run_terminal_cmd shell execution (0f3346d9) (@vinhnx)
- Fix command parsing for PTY shell compatibility (07aeb131) (@vinhnx)
- Improve terminal command rendering and disable streaming mode (d82156c4) (@vinhnx)
- Add terminal command follow-up message (20200f58) (@vinhnx)
- Handle tagged tool call transcripts (033e82c1) (@vinhnx)
- Handle one-based indices in tagged tool calls (022967d1) (@vinhnx)
- Parse run_terminal_cmd rust blocks in textual tool calls (76da1941) (@vinhnx)
- Fix run_terminal_cmd structured parsing edge cases (3611efd9) (@vinhnx)
- Auto-approve run_terminal_cmd and clean command values (499daa32) (@vinhnx)
- Render textual tool requests in panels (cf752d15) (@vinhnx)
- Tighten textual tool rendering (0f8f0a1f) (@vinhnx)
- Fix Windows command parsing and quoting (9f7f68bc) (@vinhnx)
- Fix Windows shell quoting and add shell-specific escaping (b0d296b9) (@vinhnx)
- Clarify CommandTool doc comment (6fb5ca6d) (@vinhnx)
- Enhance PTY tools with interactive session support (f850a2c5) (@vinhnx)
- Use sh fallback when default shell is unavailable (e3fcdbbe) (@vinhnx)
- Updates (7dbc60dc) (@vinhnx)
- Fix run_terminal_cmd compatibility for ACP (69970855) (@vinhnx)
- Support ACP terminal embedding for run_terminal_cmd (6632f196) (@vinhnx)
- Add Claude Haiku 4.5 anthropic model and update picker (72f2909b) (@vinhnx)
- Scope reasoning support to Claude Opus and Sonnet (ccd47ff2) (@vinhnx)
- Allow Claude Haiku 4.5 reasoning and refresh release date (b303bedc) (@vinhnx)
- Update reasoning effort labels (040a8001) (@vinhnx)
- Align init command with agents.md specification (5bc47073) (@vinhnx)
- Remove redundant workspace rule shim (07483e4f) (@vinhnx)
- Silence mcp-types dead code warnings (147d95bf) (@vinhnx)
- Refine AGENTS.md discovery and prompt context (675f10c9) (@vinhnx)
- Update default OpenAI model (23f4cd2b) (@vinhnx)
- Align CLI API key inference with default provider (3a39d937) (@vinhnx)
- Restore Gemini defaults for config provider (9c7e84ff) (@vinhnx)
- Improve OpenAI request capability gating (fd402f37) (@vinhnx)
- Fix OpenAI GPT-5 responses handling (2938d194) (@vinhnx)
- Align OpenAI responses payload with official API (9206ff1d) (@vinhnx)
- Add interactive first-run setup wizard (5475da54) (@vinhnx)
- Prompt first-run users to run init setup (9ef74726) (@vinhnx)
- Respect non-interactive flags in first-run wizard (ec03932b) (@vinhnx)
- Clarify first-run rerun instructions (df31d45b) (@vinhnx)
- Update chat placeholder to mention /init (84100e8c) (@vinhnx)
- Refine chat placeholder progression (a07b1eaf) (@vinhnx)
- Add .vtcode/index/ to .gitignore (d6269ceb) (@vinhnx)
- Update vtcode.toml configuration for API keys and model settings (5d36a33a) (@vinhnx)
- Fix Zed read_file tool schema for ACP (9fb9110e) (@vinhnx)
- Improve Zed ACP tool schemas (acd75145) (@vinhnx)
- Integrate portable-pty into terminal command execution (792b1020) (@vinhnx)
- Document VT Code-specific status line guidance (d0330a6b) (@vinhnx)
- Add JSON output option for ask command (9ea5e760) (@vinhnx)
- Respect /clear usage and index linked directories (a4da5d05) (@vinhnx)
- Enhance slash commands for MCP and workspace management (b001974a) (@vinhnx)
- Restrict command usage with session resume in CLI (0b8db06d) (@vinhnx)
- Add documentation for resuming sessions in CLI (dd3c5b47) (@vinhnx)
- Add .vtcode/index/ to .gitignore (22b2aeff) (@vinhnx)
- Update configuration files with enhanced settings and improved documentation (62bfc2fd) (@vinhnx)
- Fix OpenAI GPT-5 reasoning configuration (ecdc62c4) (@vinhnx)
- Enable streaming reasoning traces for OpenAI (25a1ac8a) (@vinhnx)
- Add comprehensive project analysis report (65b835e5) (@vinhnx)
- Add refactoring checklist to analysis report (65e37b85) (@vinhnx)
- Update default model to GPT-5 and adjust router model references (49c397b1) (@vinhnx)
- Enhance session resume functionality with workspace metadata handling (7dbeb850) (@vinhnx)
- Refactor command allow/deny lists for improved readability (d76d3baa) (@vinhnx)
- Refine ACP client registration (e52afccd) (@vinhnx)
- Extract ACP client helpers into reusable crate (f3d763c4) (@vinhnx)
- Improve checkpoint retention and configuration (b33ed1d8) (@vinhnx)
- Canonicalize snapshot paths before stripping workspace prefix (0c4db310) (@vinhnx)
- Add reasoning field to Message struct and update related implementations (b33a54a3) (@vinhnx)
- Refactor provider configuration helpers (3fbf58b1) (@vinhnx)
- Refactor provider configuration to switch from OpenAI to OpenRouter (00e881d4) (@vinhnx)
- Resolve provider config conflicts with main (d89cb97f) (@vinhnx)
- Refactor run_status_line_command to improve stdout handling and timeout management (dd8bc71b) (@vinhnx)
- Condense inline session header (fe95286c) (@vinhnx)
- Tighten inline header highlights (b8cf1508) (@vinhnx)
- Remove languages summary from inline header (4275cbfe) (@vinhnx)
- Add reasoning field to provider messages (c79d0d22) (@vinhnx)
- Stop truncating slash command header preview (ffeab1fe) (@vinhnx)
- Revise instruction memory guide for VT Code (76864109) (@vinhnx)
- Add interactive mode reference documentation (27621d14) (@vinhnx)
- Add print mode flag for single-shot CLI queries (5bf2a1c3) (@vinhnx)
- Enforce print mode restrictions and clean up redundant reasoning fields (2a7ea67b) (@vinhnx)
- Refactor import order and simplify context handling in status line command (1c238957) (@vinhnx)
- Simplify wait handling in run_status_line_command by removing mutable reference (fb083959) (@vinhnx)
- Refactor tool output display, enhance tool permission handling, and update CLI documentation (9d1e6dd8) (@vinhnx)
- Refactor unified turn loop into modular helpers (81b803f8) (@vinhnx)
- Update agent configuration to use OpenRouter provider and adjust model settings (1e73aaf7) (@vinhnx)
- Add process hardening safeguards (df42c3e9) (@vinhnx)
- Load dotenv before process hardening (42fe197e) (@vinhnx)
- Integrate ansi-to-tui for inline ANSI rendering (fbe6bd59) (@vinhnx)
- Support run alias for terminal tool detection (e0a52413) (@vinhnx)
- Add input scroll offset management and adjust rendering logic (ef9ed89c) (@vinhnx)
- Improve inline TUI terminal mode handling (db389c70) (@vinhnx)
- Enforce exec policy for terminal commands (5c2e84c2) (@vinhnx)
- Harden execpolicy path checks against symlinks (6929d684) (@vinhnx)
- Enhance TUI hitl modals and block styling (0b37c4df) (@vinhnx)
- Enhance HITL list filtering feedback (cb89be33) (@vinhnx)
- Improve PTY and code fence handling in inline session (d9076aef) (@vinhnx)
- Hide empty PTY blocks until output appears (e5e931c7) (@vinhnx)
- Add function call block parsing and enhance argument handling (35454cde) (@vinhnx)
- Add OpenRouter free model entries (defce92e) (@vinhnx)
- Prevent repeated tool loops and streamline reasoning render (f37d4eff) (@vinhnx)
- Remove demo image from README (a720fe0b) (@vinhnx)
- Refactor TUI session modal into modular components (cfa0c45e) (@vinhnx)
- Handle textual run_terminal_cmd tool invocations (a0733c34) (@vinhnx)
- Use JSON5 parser for textual run_terminal_cmd (8d277afa) (@vinhnx)
- Require fenced run_terminal_cmd text blocks (f3ecc90d) (@vinhnx)
- Fix transcript padding when viewport is small (a432b140) (@vinhnx)
- Switch token budgeting to Hugging Face tokenizers (c3de1189) (@vinhnx)
- Document crates.io categories in crate docs (1299b844) (@vinhnx)
- Refactor OpenAIProvider to support Responses API and add largest files mode to FileOpsTool (c8507155) (@vinhnx)
- Refactor tool output rendering and enhance model selection logic (ed250171) (@vinhnx)
- Replace pulldown markdown parser with comrak (436e2ac1) (@vinhnx)
- Address markdown parser warnings (ce389831) (@vinhnx)
- Enhance slash palette search ranking (83643d15) (@vinhnx)
- Fix slash palette command range scanning (a87981a7) (@vinhnx)
- Add tree-sitter bash integration (afd25765) (@vinhnx)
- Add Codex-style inline config overrides (58b0f5d7) (@vinhnx)
- Use Ratatui provider selector in setup wizard (58330f8f) (@vinhnx)
- Ensure provider selector restores terminal state on errors (adbc294a) (@vinhnx)
- Handle Ctrl+C in provider selector (34eb5f88) (@vinhnx)
- Improve setup selection UX and update OpenAI defaults (a724a457) (@vinhnx)
- Refine unified turn loop context and tool handling (b022d294) (@vinhnx)
- Add ACP adapter permission integration tests (ea0c9e67) (@vinhnx)
- Deduplicate ACP adapter definition (46258785) (@vinhnx)
- Expose ACP modules for integration tests (d9373e65) (@vinhnx)
- Improve chat input with multiline textarea (b488635b) (@vinhnx)
- Adjust chat input height to grow dynamically (ce650be4) (@vinhnx)
- Add chat input history navigation (92a792cc) (@vinhnx)
- Gate chat history navigation behind modifiers (33f3abd2) (@vinhnx)
- Refactor curator parameter in run_single_agent_loop_unified function (577d2e90) (@vinhnx)
- Fix input scroll offset references in Session implementation (b1fe9f6b) (@vinhnx)
- Align exec mode events with codex schema (946efdc6) (@vinhnx)
- Ensure exec mode runs fully autonomously (f04d334e) (@vinhnx)
- Add core interface layer for turn driving and ACP (b94f6e21) (@vinhnx)
- Integrate perg fallback for rp_search (240dff53) (@vinhnx)
- Prefer rp_search tool over shell grep (061beeaa) (@vinhnx)
- Canonicalize rp_search defaults (d9d32fef) (@vinhnx)
- Remove legacy grep_search alias (59d79444) (@vinhnx)
- Rename rp_search tool to grep_file (7a55cecd) (@vinhnx)
- Fix perg fallback parsing for Windows paths (1e820a77) (@vinhnx)
- Fix perg output parsing for Windows paths (b435d142) (@vinhnx)
- Add MCP timeouts and experimental RMCP toggle (acdd7441) (@vinhnx)
- Handle potential loop detection responses (9902ee27) (@vinhnx)
- Fix FreeBSD build by gating RLIMIT exit code (a3d8ee0b) (@vinhnx)
- Remove duplicate Zed ACP adapter implementation (bc743feb) (@vinhnx)
- Replace vt100 parser with avt (e9e7ed04) (@vinhnx)
- Deduplicate Zed ACP adapter definition (331adf22) (@vinhnx)
- Preserve UTF-8 boundaries in PTY reader (82a1228e) (@vinhnx)
- Improve custom prompt listing messaging (7cd73b97) (@vinhnx)
- Document custom prompt workflows (0b023eca) (@vinhnx)
- Add built-in vtcode custom prompt (8b1ceef7) (@vinhnx)
- Allow custom prompt slash command to submit prompt (2ea7a858) (@vinhnx)
- Bundle built-in custom prompt with vtcode-core (6d1a468c) (@vinhnx)
- Handle slash command tool execution (97426813) (@vinhnx)
- Refactor SlashPalette to support custom prompts and enhance suggestion handling (c4e03dc7) (@vinhnx)
- Restore interactive workspace trust prompt (24d74cf2) (@vinhnx)
- Fix OpenRouter build outputs to restore runtime deps (1433d4f1) (@vinhnx)
- Harden Windows process startup mitigations (22da3c4f) (@vinhnx)
- Add CI guard for oversized tracked files (b5719a37) (@vinhnx)
- Version large file allowlist for CI guard (c811db41) (@vinhnx)
- Normalize large file allowlist matching (3a755207) (@vinhnx)
- Add per-pattern thresholds to large file checker (ca6891aa) (@vinhnx)
- Fix OpenRouter metadata warnings and hardening imports (fb0e5d80) (@vinhnx)
- Embed OpenRouter metadata for standalone builds (92fddb42) (@vinhnx)
- Handle unsupported Windows mitigations (27d0b102) (@vinhnx)
- Enable Windows SystemServices module (7348adc1) (@vinhnx)
- Fix tool call delta recursion index handling (96ce5163) (@vinhnx)
- Change badge styles to flat-square in README (75fbb347) (@vinhnx)
- Refactor grep search module: rename and update references to grep_file (e69f761b) (@vinhnx)
- Format code for better readability in runloop and session modules (9879297e) (@vinhnx)
- Fix path resolution for default .vtcode.toml template (48f279db) (@vinhnx)
- Update (3fa734bb) (@vinhnx)
- Document vtcode-commons reference adapters (55554a61) (@vinhnx)
- Fix error reporter and telemetry trait bounds (cd2c75d7) (@vinhnx)
- Constrain memory telemetry events to be Send + Sync (eec413fc) (@vinhnx)
- Support queuing chat inputs (7bf14d65) (@vinhnx)
- Refactor tool output rendering and improve message styles (0ffaaf93) (@vinhnx)
- Restore newline entry while reserving queue shortcut (aa8ef2eb) (@vinhnx)
- Adjust inline editor shortcuts tests (0e993c4e) (@vinhnx)
- Preserve tool outputs in Harmony conversations (106d85d2) (@vinhnx)
- Detect Ollama models when inferring providers (1a17b0d2) (@vinhnx)
- Fix harmony tool call content extraction (b22a76c1) (@vinhnx)
- Ensure fallback responses update streamed output (68502191) (@vinhnx)
- Normalize Harmony tool call argument JSON (ac2b38ab) (@vinhnx)
- Include Harmony tool definitions in prompts (8034d5e8) (@vinhnx)
- Disable dependencies check in CI workflow (6758a53f) (@vinhnx)
- Update tool display (bf840416) (@vinhnguyenxuan-ct)
- Handle tool history for OpenRouter gpt-oss (9011d887) (@vinhnx)
- Fix harmony model detection for GPT-OSS variants (0cd3c0f5) (@vinhnx)
- Guard theme selection during model picker (e35c86d7) (@vinhnx)
- Improve condensed shell output messaging (145871b1) (@vinhnx)
- Allow vtcode custom prompt positional task argument (466768c2) (@vinhnx)
- Show queued inputs at bottom of transcript (71d2993f) (@vinhnx)
- Limit queued input overlay length (ac5684f7) (@vinhnx)
- Mark queued input updates for redraw (5e9c573b) (@vinhnx)
- Add slash command to generate AGENTS instructions (6d755ec2) (@vinhnx)
- Improve AGENTS.md generator output (2ca03dbd) (@vinhnx)
- Enhance interactive setup list UI (33c5b96a) (@vinhnx)
- Refine ratatui interactive selection (f20fc177) (@vinhnx)
- Polish tool output ratatui panels (ce51f78f) (@vinhnx)
- Refine terminal command panel layout (61ded116) (@vinhnx)
- Widen plan and command panels (e113199e) (@vinhnx)
- Revert "Widen plan and command panels" (5ab5e410) (@vinhnguyenxuan-ct)
- Revert "Refine terminal command panel layout" (22ccde25) (@vinhnguyenxuan-ct)
- Restore generate-agent-file slash command dispatch (e6dbea01) (@vinhnx)
- Refactor output rendering and improve command security in execution policy (34934158) (@vinhnguyenxuan-ct)
- Add cancellation messages and support in agent loop and streaming response (dccf80eb) (@vinhnguyenxuan-ct)
- Refactor terminal command panel rendering to simplify color handling and improve stdout/stderr processing (1b30ed82) (@vinhnx)
- Add sandbox runtime integration for bash tool (be3c3b6a) (@vinhnx)
- Allow network commands when sandboxing bash tool (477e8d34) (@vinhnx)
- Fix sandbox runtime detection context string (44cc722f) (@vinhnx)
- Allow network commands in sandbox (8b7fbf2e) (@vinhnx)
- Integrate Anthropic sandbox runtime for enhanced security in command execution (99a7b6b1) (@vinhnx)
- Adjust MCP config for 2024-11 schema (66019bba) (@vinhnx)
- Rearrange badges in README.md (26e05463) (@vinhnx)
- Document vtcode-config migration path (c53a33ef) (@vinhnx)
- Scaffold vtcode-config crate defaults (2b9b8692) (@vinhnx)
- Move config loader into vtcode-config (96d0b53f) (@vinhnx)
- Fix vtcode config defaults helper and openrouter metadata (9f7adaa7) (@vinhnx)
- Use provider module for config defaults test helper (3a38f648) (@vinhnx)
- Restore project config lookup in loader (4b299f90) (@vinhnx)
- Simplify vtcode-core loader to re-export shared module (25c42273) (@vinhnx)
- Add loader tests and crate docs for vtcode-config (511ded79) (@vinhnx)
- Gate vtcode-config bootstrap behind feature flag (eea6c406) (@vinhnx)
- Add schema export helpers to vtcode-config (67bd3d1b) (@vinhnx)
- Add vtcode-config usage example and docs (e110de13) (@vinhnx)
- Run loader defaults tests serially (01762c07) (@vinhnx)
- Add MCP resource and prompt support (61c16301) (@vinhnx)
- Handle telemetry deltas under release builds (c633b8ab) (@vinhnx)
- Enhance harmony tool call parsing and output handling; add tests for new functionality (4b0234f8) (@vinhnx)
- Update tool policies to ensure explicit prompts for sensitive operations (c8fa4dcb) (@vinhnx)
- Add pluggable storage and filters to vtcode-indexer (c9872da3) (@vinhnx)
- Document vtcode-indexer customization hooks (17a25c7c) (@vinhnx)
- Outline vtcode-bash-runner extraction strategy (4b907f23) (@vinhnx)
- Scaffold vtcode-bash-runner crate (9085a7f3) (@vinhnx)
- Document vtcode-bash-runner and add dry-run example (06bd047d) (@vinhnx)
- Outline vtcode-exec-events extraction strategy (c919f054) (@vinhnx)
- Scaffold vtcode-exec-events crate (9b20f4f4) (@vinhnx)
- Document vtcode-exec-events telemetry (18bfac6e) (@vinhnx)
- Add vtcode-exec-events JSONL example (0967bf5b) (@vinhnx)
- Document release plan for extracted crates (c3b03b3c) (@vinhnx)
- Prepare extracted crates for 0.1.0 releases (82d27cf8) (@vinhnx)
- Document dry-run validations for extracted crates (733635b1) (@vinhnx)
- Document sequential publish schedule for extracted crates (cd6a8cd3) (@vinhnx)
- Enhance vtcode-exec-events integration surface (ffa04b1b) (@vinhnx)
- Expand vtcode-bash-runner features and telemetry (35de449c) (@vinhnx)
- Harden workspace boundary checks (65ed9ee3) (@vinhnx)
- Emit failed status for non-zero exits (113e2eda) (@vinhnx)
- Canonically validate mutating command targets (ffd19f0d) (@vinhnx)
- Canonicalize symlink mutation targets (a4fc6d27) (@vinhnx)
- Fix link to MCP Setup in README (9e0c897d) (@vinhnx)
- Restore GPT-4o tokenizer mapping (d4dbac50) (@vinhnx)
- Add free-tier knobs (sleep/backoff/retries); MBPP/HE/SWE-lite-dry + Makefile (c9babd50) (@vinhnx)
- Refactor benchmark scripts for improved readability and functionality (e13e81b6) (@vinhnx)
- Implement context engineering enhancements: Phase 1 & 2 (d6279392) (@vinhnx)
- Add benchmark comparison charts and scripts (63915651) (@vinhnx)
- Fix table formatting in README.md (3f8a8351) (@vinhnx)
- Revise benchmarks and remove usage examples in README (dbe2f087) (@vinhnx)
- Clarify benchmark evaluation statement in README (ac5bee0c) (@vinhnx)
- Update benchmark documentation with corrected pricing and model comparisons (9800a218) (@vinhnx)
- Remove benchmarks section from README (9b5cf6ec) (@vinhnx)
- Support interactive MCP elicitation handling (8e202ea0) (@vinhnx)
- Remove tests (023a47de) (@vinhnx)
- Enhance Ollama tool support (64ad3857) (@vinhnx)
- Allow auto tool choice and bool reasoning for Ollama (4e3b8605) (@vinhnx)
- Allow auto tool choice and enable reasoning flag (cdf30031) (@vinhnx)
- Add configurable base URLs for hosted providers (a753fb19) (@vinhnx)
- Add Accept-Language header for Z.AI requests (b81efb9c) (@vinhnx)
- Restore Ollama base URL override (bf30e47c) (@vinhnx)
- Refactor Ollama model handling to differentiate between cloud and local models regarding API key usage (d2f02c38) (@vinhnx)
- Add Ollama API key support in ApiKeySources and implement retrieval functions (12537ded) (@vinhnx)
- Allow local Ollama runs without API keys (b927835d) (@vinhnx)
- Handle raw string tool arguments from Ollama (7da23e1f) (@vinhnx)
- Respect disabled tool choice in Ollama payload (587fb944) (@vinhnx)
- Default missing Ollama tool arguments to empty object (cbb18a86) (@vinhnx)
- Add benchmark reports and update package versions (42498993) (@vinhnx)
- Improve ast-grep search formatting and limits (e3fda3ea) (@vinhnx)
- Handle nested ast-grep lines object (a479d7e5) (@vinhnx)
- Handle string ast-grep lines (301d77f0) (@vinhnx)
- Fix ast-grep custom match extraction (3ca0efb5) (@vinhnx)
- Handle nested AST-grep custom results (cc204a91) (@vinhnx)
- Fix custom AST-grep metadata merge (4525b930) (@vinhnx)
- Refactor tools to support asynchronous operations (7bded590) (@vinhnx)
- Refactor async function calls for improved readability and consistency (578166ac) (@vinhnx)
- Add create_file tool and update related documentation (d05bf20f) (@vinhnx)
- Fix nested metadata merge for AST-grep (61deb95c) (@vinhnx)
- Enhance timeout detection, token budget management, and file operations (7f73c2dc) (@vinhnx)
- Enhance tool execution and output handling (325971c8) (@vinhnx)
- Add demo TUI script, basic Rust program, and comprehensive deny.toml configuration (2622cbe8) (@vinhnx)
- Add disable reasoning option to model picker (5af8be2b) (@vinhnx)
- Allow VHS demo gif in large file check (8c0bb5b8) (@vinhnx)
- Align release script with nextest checks (e89c614c) (@vinhnx)
- Prepare extracted crates for publish-ready manifests (c4d71772) (@vinhnx)
- Update QR Code image to be a clickable link (2de8edfe) (@vinhnx)
- Refactor README to remove demo duplication (29b0a09c) (@vinhnx)
- Generate docs during release automation (d223b4bf) (@vinhnx)
- Remove demo section from README (9b9e2e6e) (@vinhnx)
- Enhance security documentation and implementation for VT Code (ce7fe5f9) (@vinhnx)
- Implement self-update mechanism with downloader, installer, verifier, and rollback manager (1024cf09) (@vinhnx)
- Update README with installation instructions and description (0b0f71f1) (@vinhnx)
- Add starter VS Code extension (eadd278f) (@vinhnx)
- Enhance VS Code extension integration (0142be78) (@vinhnx)
- Expand VTCode VS Code onboarding (004c6ac1) (@vinhnx)
- Enhance VTCode extension language tooling (87950de8) (@vinhnx)
- Add CLI-aware UX and linting to VTCode extension (c7da6f35) (@vinhnx)
- Add integrated terminal and analysis commands to VS Code extension (63ca12fb) (@vinhnx)
- Document VTCode extension development and release plan (67e7d71d) (@vinhnx)
- Propagate vtcode config to extension CLI invocations (e81ebe46) (@vinhnx)
- Document VS Code dev and release workflow (3ddcec1a) (@vinhnx)
- Avoid shell execution when spawning VTCode CLI (9963759b) (@vinhnx)
- Align CLI detection cwd with command execution (217c20d4) (@vinhnx)
- Refactor code structure for improved readability and maintainability (00cdb985) (@vinhnx)
- Respect default shell for agent terminal (84c09ef2) (@vinhnx)
- Make Quick Actions view visible (856ef36e) (@vinhnx)
- Fix syntax error in release.sh script (eb9a4347) (@vinhnx)
- Revise README.md for improved clarity and structure (c209687e) (@vinhnx)
- Add LM Studio provider integration (c4ef825c) (@vinhnx)
- Expand LM Studio catalog entries (7c942e16) (@vinhnx)
- Add configurable sandbox runtime and filesystem allowlist (d622fb55) (@vinhnx)
- Replace VS Code Marketplace badge with custom icon (9bc057b0) (@vinhnx)
- Align system prompt with TigerStyle principles (00c1f4de) (@vinhnx)
- Refine core system prompt and remove 'TigerStyle' references (8cee7a1b) (@vinhnx)
- Refactor model picker module (51e7473a) (@vinhnx)
- Honor pre-tool hook ask decisions (5f917448) (@vinhnx)
- Ensure pre-tool hook allow respects deny policies (bf3b0a93) (@vinhnx)
- Add lifecycle hooks support and example configurations (aa822026) (@vinhnx)
- Update documentation for configuration precedence and getting started guide (ca590865) (@vinhnx)
- Update README.md (746565cb) (@vinhnx)
- Update lifecycle-hooks.md with Claude Code Hooks link (fd645159) (@vinhnx)
- Respect post-tool hook blocks (e202ce3b) (@vinhnx)
- Persist blocked hook context (1ba4f276) (@vinhnx)
- Enhance UI responsiveness and tool output management (c8f46aec) (@vinhnx)
- Implement TUI mode environment variable management and enhance tool permission handling (421ef599) (@vinhnx)
- Refactor code formatting for improved readability in tool_output.rs and lifecycle.rs (5d4c68ca) (@vinhnx)
- Fix lifecycle hook tests by using printf instead of echo (9facfa59) (@vinhnx)
- Skip lifecycle hook tests on non-macOS (2f0704aa) (@vinhnx)
- Add startup_timeout_ms field to MCP provider configuration (0db8a41d) (@vinhnx)
- Add GitHub Actions workflow for mdBook deployment (e614eb15) (@vinhnx)
- Update .github/workflows/mdbook.yml (86ddb8c1) (@vinhnx)
- Update .github/workflows/mdbook.yml (cefced2f) (@vinhnx)
- Updates (8a6122e3) (@vinhnx)
- Update installation instructions formatting in README (4a8a179a) (@vinhnx)
- Update CHANGELOG for version 0.35.1 (73652b86) (@vinhnx)
- Potential fix for code scanning alert no. 17: Cleartext transmission of sensitive information (c5e999b4) (@vinhnx)
- Update CHANGELOG for version 0.35.1 (683c7987) (@vinhnx)
- Potential fix for code scanning alert no. 14: Incomplete string escaping or encoding (479098db) (@vinhnx)
- Potential fix for code scanning alert no. 13: Incomplete string escaping or encoding (f21abcac) (@vinhnx)
- Potential fix for code scanning alert no. 12: Incomplete URL substring sanitization (4a44a5d4) (@vinhnx)
- Enhance npmrc validation to restrict registry to GitHub Packages and improve URL sanitization (a056b707) (@vinhnx)
- Update README.md (16e0aae6) (@vinhnx)
- Update documentation link in README.md (475d9a7e) (@vinhnx)
- Fix link to IDE Downloads in README.md (79d21fbf) (@vinhnx)
- Updates (990b98d5) (@vinhnx)
- [Finish #372] Implement vtcode_docs_map.md similar to claude_code_docs_map (9a493924) (@vinhnx)
- Fix build error (7b149aa5) (@vinhnx)
- Fix build error (be45c39b) (@vinhnx)
- Fix build error (451e2d02) (@vinhnx)
- Fix build error (7d716d08) (@vinhnx)
- Updates (e745d642) (@vinhnx)
- Add DeepWiki link to README (3425365b) (@vinhnx)
- Simplify VT Code description in README (e97019db) (@vinhnx)
- Refine patch applicator execution layers (bfc87996) (@vinhnx)
- Refactor TUI session into input and queue modules (07351299) (@vinhnx)
- Factor tool output handlers into dedicated modules (49f7ebd9) (@vinhnx)
- Streamline inline event polling (cea97118) (@vinhnx)
- Refactor terminal execution preparation (314b2bb7) (@vinhnx)
- Refactor ACP agent implementation and remove performance command (d802365a) (@vinhnx)
- Update tool policy to allow sending pty input and remove unused tui-scrollview package (75c5199e) (@vinhnx)
- Add environment variables to disable color output and implement test for ANSI code stripping (047a9483) (@vinhnx)
- Refactor ANSI code checks in pty output test for clarity (eedefe08) (@vinhnx)
- Update Zed extension manifest and release script for checksum automation (72723b88) (@vinhnx)
- Update to use core tap (cd49e7a3) (@chenrui333)
- Remove homebrew setup (26a6ac99) (@chenrui333)
- Refine MCP integration guide with spec references (0bc2ca02) (@vinhnx)
- Integrate VTCode chat participant and AI tools (83d6d5dd) (@vinhnx)
- Expose VS Code editor context to vtcode prompts (9c14c28c) (@vinhnx)
- Bridge IDE context between VS Code and CLI (b0588771) (@vinhnx)
- Refactor test data generation and add new test scripts (b17d4e0f) (@vinhnx)
- Refactor conditional statements for improved readability (2f4ec0a7) (@vinhnx)
- Updates (f011ca47) (@vinhnx)
- Updates (23c6fd95) (@vinhnx)
- Updates (bc91237d) (@vinhnx)
- Updates (ddbfe217) (@vinhnx)
- Updates (fb42b93f) (@vinhnx)
- Updates (ed1aed45) (@vinhnx)
- Add new release CIs (09466e2a) (@vinhnx)
- Fix release script (2963c907) (@vinhnx)
- Update Homebrew formula to version 0.39.13 (83fc2300) (@vinhnx)
- Fix build (cffabb07) (@vinhnx)
- Updates (509d1e1c) (@vinhnx)
- Add Codex issue review research summary (7157bee0) (@vinhnx)
- Improve tool timeout handling and caching (602c7920) (@vinhnx)
- Add tests and docs for sandbox module reuse (23c74255) (@vinhnx)
- Add language consistency tests, tool serialization stability tests, and telemetry event definitions (2ab31c26) (@vinhnx)
- Updates (682987f1) (@vinhnx)
- Refactor unified turn loop into submodules (0f73a058) (@vinhnx)
- Integrate terminal bridge with VS Code views (5332bbb1) (@vinhnx)
- Update extension (7f10a6bd) (@vinhnx)
- Updates (70f561aa) (@vinhnx)
- Restore session message rendering helpers (6a2e34fa) (@vinhnx)
- Refactor unified turn module into submodules (1eb7d308) (@vinhnx)
- Refactor slash command description handling and enhance test coverage for SlashPalette (bc1bbdba) (@vinhnx)
- Refactor turn finalization into dedicated module (0350c90f) (@vinhnx)
- Update version to 0.40.1 and format command events in package.json (98669ab0) (@vinhnx)
- Refactor slash command handling into dedicated module (0805587b) (@vinhnx)
- Refactor tool policies and enhance modal handling in palette selection (c399a043) (@vinhnx)
- Add smart summarization and MCP server configurations to vtcode.toml (dbfce1da) (@vinhnx)
- Update Homebrew formula to version 0.40.0 (4e0e1b6a) (@vinhnx)
- Remove Zed extension version and checksum update functionality from release script (17a75306) (@vinhnx)
- Update Homebrew formula to version 0.40.1 (e1104a10) (@vinhnx)
- Refactor imports in session.rs for clarity and organization (ff9e553b) (@vinhnx)
- Remove unused imports from session.rs for cleaner code (0332ebe6) (@vinhnx)
- Refactor slash command handling into dedicated module (22b23f1e) (@vinhnx)
- Refactor tool policies and enhance modal handling in palette selection (f51a585b) (@vinhnx)
- Add smart summarization and MCP server configurations to vtcode.toml (a4ce0882) (@vinhnx)
- Update Homebrew formula to version 0.40.0 (66a7a493) (@vinhnx)
- Remove Zed extension version and checksum update functionality from release script (b0551f50) (@vinhnx)
- Update Homebrew formula to version 0.40.1 (8c22c856) (@vinhnx)
- Remove CI badge from README (1c97e764) (@vinhnx)
- Update GitHub badge label in README.md (fe2d2e42) (@vinhnx)
- Add Moonshot Kimi K2 Thinking model (af1318a5) (@vinhnx)
- Support Moonshot K2 thinking reasoning responses (388a04c2) (@vinhnx)
- Add schema validation to MCP elicitation handshake (200a3c23) (@vinhnx)
- Refine MCP HTTP configuration and logging (3ffa73b3) (@vinhnx)
- Refactor update check documentation and implementation (014355ef) (@vinhnx)
- Enhance MCP initialization logging for better debugging (f61f23bc) (@vinhnx)
- Update tool policies and configuration for improved agent performance (2b65a5f5) (@vinhnx)
- Add early return for MCP tools in render_tool_output function (73934a7f) (@vinhnx)
- Remove Inspect variant from ThemePaletteMode and related references (b709b7f4) (@vinhnx)
- Refactor Moonshot provider to support Kimi K2 Thinking Heavy model and enhance reasoning parameters (93ea90bc) (@vinhnx)
- Update Homebrew formula to version 0.42.9 (f4891771) (@vinhnx)
- Update Homebrew formula to version 0.42.11 (004a5c99) (@vinhnx)
- Refactor system prompts and improve operational guidelines (b0a0cff4) (@vinhnx)
- Update Homebrew formula to version 0.42.12 (70198ebb) (@vinhnx)
- Update Homebrew formula to version 0.42.13 (6c5fffab) (@vinhnx)
- Update Homebrew formula to version 0.42.14 (5fe11de0) (@vinhnx)
- Update Homebrew formula to version 0.42.15 (d9173e10) (@vinhnx)
- Revert "feat: add configurable LLM generation parameters in vtcode.toml" (a6310d5c) (@vinhnx)
- Reapply "feat: add configurable LLM generation parameters in vtcode.toml" (00b2ff51) (@vinhnx)
- Revert "refactor: update tool policies and improve MCP tool handling" (a49c0350) (@vinhnx)
- Refactor network tools: Remove CurlTool, introduce WebFetchTool (de30d798) (@vinhnx)
- Update Homebrew formula to version 0.42.16 (7f80ccf6) (@vinhnx)
- More proper Anthropic Claude support (14ce79f6) (@vinhnx)
- Update Homebrew formula to version 0.42.17 (6c0effad) (@vinhnx)
- Step 6: Add integration test framework for MCP code execution (0a0fe3b9) (@vinhnx)
- Step 6: Add comprehensive documentation for code execution (10715534) (@vinhnx)
- Step 6: Add completion summary - all validation criteria met (1d3c8577) (@vinhnx)
- Add CODE_EXECUTION_QUICK_START.md - Fast reference guide (b2cf9d15) (@vinhnx)
- Add complete implementation status documentation for MCP code execution (f7d26cdd) (@vinhnx)
- Remove verbose static content from agent prompts (335a6873) (@vinhnx)
- Revert "optimize: remove verbose static content from agent prompts" (11714182) (@vinhnx)
- Refactor execution policy and remove git_diff tool (ac762062) (@vinhnx)
- Eliminate git_diff tool from available tools and policies (577ddaa0) (@vinhnx)
- Add small model tier support for efficient operations (be23c3eb) (@vinhnx)
- Update tool policies to change input permissions and remove unused methods (26bf49b8) (@vinhnx)
- Add comprehensive git command execution documentation and quick reference (cc44163c) (@vinhnx)
- Add comprehensive review documents for VT Code's VSCode extension improvements (2a33df33) (@vinhnx)
- Update Homebrew formula to version 0.42.19 (7c85b337) (@vinhnx)
- Updates (01d94208) (@vinhnx)
- Update Homebrew formula to version 0.42.20 (c919bbb8) (@vinhnx)
- Updates (63134720) (@vinhnx)
- Update Homebrew formula to version 0.43.1 (31dbbd50) (@vinhnx)
- Remove duplicate installation commands from README (fff02dd6) (@vinhnx)
- Add command builder, metrics, and timeout handling (fae1fbf6) (@vinhnx)
- Add comprehensive documentation and release notes for VTCode Zed Extension (14b0f2d9) (@vinhnx)
- Refactor installation documentation and scripts for VT Code (a7c8c112) (@vinhnx)
- Update Homebrew formula to version 0.43.2 (c295fd42) (@vinhnx)
- Remove vtcode installation instructions from README (73c3eb07) (@vinhnx)
- Refactor color parsing and styling utilities (4ba30733) (@vinhnx)
- Update Homebrew formula to version 0.43.3 (0d8ba62b) (@vinhnx)
- Update README.md (0dffa251) (@vinhnx)
- Implement LRU Cache and Performance Optimizations (65c5e4db) (@vinhnx)
- Enhance grep_file Tool and Documentation (46698574) (@vinhnx)
- Update Homebrew formula to version 0.43.4 (3c68ed94) (@vinhnx)
- Refactor context management and remove context compression features (3e0b869f) (@vinhnx)
- Update VHS showcase and demo scripts to remove slash command references (fc0aeda4) (@vinhnx)
- Refactor context management and error recovery (bdaf0a86) (@vinhnx)
- Phase 5.7: Integrate ToolPermissionCache for tool-level approval grants (f1b3279d) (@vinhnx)
- Phase 6.3-6.4: Provider Optimization and Advanced Context Pruning (27ccf504) (@vinhnx)
- Phase 6.5.3 & 6.6.2: TUI Metrics Panel and ContextPruner Integration (66b66779) (@vinhnx)
- Phase 6.6.3: TUI Status Line Integration with Context Efficiency Metrics (d9b39b4b) (@vinhnx)
- Refactor and clean up code across multiple modules (2e532067) (@vinhnx)
- Enhance shell resolution to ensure infallible fallback and streamline command execution (9a53329e) (@vinhnx)
- Update run script for clearer build mode messages and adjust default terminal timeout settings (55ef3191) (@vinhnx)
- Update Homebrew formula to version 0.43.5 (b24d660d) (@vinhnx)
- Fix tests failure (ea8f37f0) (@vinhnx)
- Update Homebrew formula to version 0.43.6 (38b29b3f) (@vinhnx)
- Updates (33ec37c3) (@vinhnx)
- Update TODO (77c15deb) (@vinhnx)
- Fix build (5ea42ccd) (@vinhnx)
- Update Homebrew formula to version 0.43.11 (cdb307b7) (@vinhnx)
- Refactor timeout handling and add streaming progress tracking across LLM providers (fb3486fa) (@vinhnx)
- Add support for anstyle-git and enhance diff rendering; update LLM provider settings (7afacc13) (@vinhnx)
- Update Homebrew formula to version 0.43.15 (e7427129) (@vinhnx)
- Update BuyMeACoffee badge color in README (430b4bd1) (@vinhnx)
- Add verbosity control and enhance reasoning effort levels in LLM configuration (f0c1d021) (@vinhnx)
- Refactor function handling across providers to ensure safe access and improve error handling; update system prompts for enhanced persistence and instruction-following; modify tool descriptions for clarity; switch default LLM provider to OpenAI and update model configurations. (8465d8f4) (@vinhnx)
- Update Homebrew formula to version 0.43.17 (af42a73f) (@vinhnx)
- VS Code Extension - View containers in Secondary Side Bar (30ca73a9) (@vinhnx)
- Response API prompt_cache_retention (58bd9db4) (@vinhnx)
- Get_errors tool (0ba5a718) (@vinhnx)
- Enhance Web Fetch Tool with dynamic configuration and default headers (dee6921b) (@vinhnx)
- Add output_format support across various components and providers (a47b1399) (@vinhnx)
- Refactor reasoning parameters and update model configurations for GPT-5.1 and Claude 4.x models (925a0de8) (@vinhnx)
- Add new models and update constants for Claude and GPT variants (e7e54b81) (@vinhnx)
- Update Homebrew formula to version 0.44.1 (e155faf8) (@vinhnx)
- Remove BuyMeACoffee badge (90738515) (@vinhnx)
- Restore previous run_loop.rs baseline before pipeline extraction (f1fb4445) (@vinhnx)
- Implement tool execution and turn processing modules; add retry logic for PTY commands; update configuration for new LLM provider. (56737b41) (@vinhnx)
- Refactor LLM response processing and update tests for verbosity level (ae95005f) (@vinhnx)
- Refactor command validation to require confirmation for destructive operations and enhance audit logging (11beb7bf) (@vinhnx)
- Update Homebrew formula to version 0.45.0 (9aae9190) (@vinhnx)
- Implement token budget management and truncation features across tools and commands (28879b4d) (@vinhnx)
- Refactor token budget management and enhance PTY session handling; implement max_tokens truncation and improve command security configurations. (b3010931) (@vinhnx)
- Update Homebrew formula to version 0.45.2 (5d2b7ddb) (@vinhnx)
- Apply cargo fmt (478483bf) (@vinhnx)
- Update Homebrew formula to version 0.45.3 (028c918b) (@vinhnx)
- Update Homebrew formula to version 0.45.4 (dbb655f3) (@vinhnx)
- Add better diagnostics for 'command not found' errors (exit code 127) (4e6b2336) (@vinhnx)
- Deprecate RUN_COMMAND in favor of RUN_PTY_CMD and remove related code; update dependencies and clean up unused imports. (3c572956) (@vinhnx)
- Phase 1: Implement context engineering & output curation in system prompt (87de9fe1) (@vinhnx)
- Phase 2: Implement multi-LLM compatibility in system prompt (c097a5f5) (@vinhnx)
- Phase 3 Research: System Prompt Optimization Planning (e011fa3b) (@vinhnx)
- Add Phase 3 Executive Handoff document (f4cfd193) (@vinhnx)
- Add Phase 3 Research Outcome: Complete Summary Document (6fbbc27d) (@vinhnx)
- Phase 3: Complete Team Execution Package (485383de) (@vinhnx)
- Enhance error handling and output formatting (3fd53b42) (@vinhnx)
- Update Homebrew formula to version 0.45.5 (15d786b7) (@vinhnx)
- Add tick and frame rate configuration to CLI; improve mutex handling; implement full TUI with event-driven architecture; enhance versioning with git info; refactor input handling and terminal management. (b62d7a91) (@vinhnx)
- Update Homebrew formula to version 0.45.6 (26b7014d) (@vinhnx)
- Phase 1: Complete MCP module exports and fix schema validation (e347d095) (@vinhnx)
- Add comprehensive MCP Phase 1 documentation (497da038) (@vinhnx)
- Add MCP review outcome report - Phase 1 complete (8b7890ff) (@vinhnx)
- Phase 2.1: Transport integration - eliminate duplicate code (fc6fe89d) (@vinhnx)
- Phase 2: Full JSON Schema 2020-12 validation implementation (a0d1aea3) (@vinhnx)
- Add Phase 2 completion report (2e886fd0) (@vinhnx)
- Add comprehensive session summary for MCP review and Phase 2 work (51c71319) (@vinhnx)
- Add MCP documentation index and navigation guide (9d39c80d) (@vinhnx)
- Phase 3.1: Error code system - add MCP_E{code} error identification (d1099d11) (@vinhnx)
- Add Phase 3.1 error code system documentation (e085a385) (@vinhnx)
- Improve MCP handling (7b5cd2bf) (@vinhnx)
- Update Homebrew formula to version 0.46.0 (81098b44) (@vinhnx)
- Refactor permission logging, enhance task outcome handling, and improve command validation; update LLM provider configurations and streamline UI event handling. (bfed8450) (@vinhnx)
- Replace static spinner icon with contextual agent message (3ad1edd9) (@vinhnx)
- Change reasoning prefix from 'Thinking:' to 'I'm thinking:' for first-person consistency (280c3177) (@vinhnx)
- Disable ANSI color codes in PTY command output (8ede8406) (@vinhnx)
- Add PTY ANSI handling tests and documentation (49fee678) (@vinhnx)
- Refactor code for improved readability and consistency; remove unnecessary whitespace and streamline function signatures across multiple files. (5f29c82c) (@vinhnx)
- Fix validation error handling in harmony args conversion (848d6f1d) (@vinhnx)
- Refactor code structure for improved readability and maintainability (f5b8f126) (@vinhnx)
- Update Homebrew formula to version 0.47.4 (3bdac872) (@vinhnx)
- Refactor linting output parsing, enhance project analysis, add ClaudeOpus 4.5 model, improve PTY command execution, and update UI styles for tool output. (b1b56072) (@vinhnx)
- Update Homebrew formula to version 0.47.5 (c80fe09a) (@vinhnx)
- Remove redundant closures (7aa996d7) (@vinhnx)
- Fix clippy warnings (assign_op, unnecessary_cast, collapsible_if, const thread_local) (4127b96e) (@vinhnx)
- Refactor code for improved readability and performance; replace `to_string()` with `into_owned()`, use `clamp()` for value constraints, and optimize iterators in various modules. (cd64b674) (@vinhnx)
- Refactor string allocations to use `into()` for improved performance and readability (de00d25c) (@vinhnx)
- Refactor string handling to use `to_owned()` instead of `to_string()` for improved performance and consistency across the codebase. (17ffccd1) (@vinhnx)
- Refactor string handling across the codebase to consistently use `to_owned()` instead of `to_string()`, enhancing performance and maintaining uniformity in string allocations. (e9b83b60) (@vinhnx)
- Sync embedded docs map with canonical docs (663feb84) (@vinhnx)
- Refactor tool execution and caching for improved performance and memory efficiency (fee2e192) (@vinhnx)
- Enhance caching mechanisms with Arc support to reduce cloning overhead (a265716f) (@vinhnx)
- Refactor code for improved readability and performance; optimize message rebuilding, error handling, and ANSI code stripping. (663d6859) (@vinhnx)
- Add tests and functionality to strip markdown formatting from command strings (3e69419a) (@vinhnx)
- Refactor timestamp handling and improve unicode processing (64c592ed) (@vinhnx)
- Implement unified symbol extraction framework and enhance terminal capabilities detection (92f506c2) (@vinhnx)
- Refactor caching logic in SmartResultCache for improved performance and clarity; update TreeSitterAnalyzer to simplify parsing and symbol extraction; enhance ANSI utilities and input handling; clean up terminal capabilities checks; improve unicode monitoring and testing; adjust LRU cache handling for better stats management. (23f7ee4e) (@vinhnx)
- Refactor result cache to use unified cache structure; optimize symbol extraction with cached node kinds; improve session listing performance with parallel processing. (eb106425) (@vinhnx)
- Optimize memory allocation and improve performance across various modules (8e9ffbc8) (@vinhnx)
- Refactor ANSI stripping functions for improved UTF-8 handling and add tests for unicode preservation (453aeb16) (@vinhnx)
- Refactor test utilities for improved organization and clarity (d0d9733c) (@vinhnx)
- Update Homebrew formula to version 0.47.6 (09efab56) (@vinhnx)
- Finishes #533 Add DeepSeek V3.2 Models (9ab69a4c) (@vinhnx)
- Update Homebrew formula to version 0.47.7 (7d5f7f04) (@vinhnx)
- Add YouTube video link to README (f913fcd2) (@vinhnx)
- Sync Claude annotations from my-claude-code (f52d80cc)
- Migrate caching implementation to UnifiedCache and deprecate LruCache; update related modules and configurations. (72234225) (@vinhnx)
- Update Homebrew formula to version 0.47.9 (afd01921) (@vinhnx)
- Refactor context management and improve loop detection; optimize token usage warnings and output formatting (d4f138ef) (@vinhnx)
- Update Homebrew formula to version 0.47.12 (4ddc3660) (@vinhnx)
- Update Homebrew formula to version 0.47.13 (47206de2) (@vinhnx)
- Add Smithery badge (e91b1140) (@gurdasnijor)
- Update Homebrew formula to version 0.47.15 (29cd0ae3) (@vinhnx)
- Update Homebrew formula to version 0.47.16 (44ce11b5) (@vinhnx)
- Update Homebrew formula to version 0.48.0 (32840573) (@vinhnx)
- Update Homebrew formula to version 0.48.2 (3cec9c06) (@vinhnx)
- Add logging functionality and improve code search descriptions (70201246) (@vinhnx)
- Enhance final response rules to prevent code dumping and improve output clarity (1cfe9d2a) (@vinhnx)
- Update Homebrew formula to version 0.48.3 (86571633) (@vinhnx)
- Update Homebrew formula to version 0.49.1 (ca51d3e1) (@vinhnx)
- Refactor command range detection to terminate on whitespace and update `execute_code` and `skill` tool policies to prompt. (55f7ee54) (@vinhnx)
- Refactor parallel execution planner and result collector; enhance tool registry with caching; add CLI tool templates and instructions (1aa105d2) (@vinhnx)
- Refactor skill command handling to utilize EnhancedSkillLoader, improving skill discovery and loading processes. Update CLI tool handling to differentiate between traditional skills and CLI tools across various command actions. (dd4df2a6) (@vinhnx)
- Update Homebrew formula to version 0.49.5 (98244bc5) (@vinhnx)
- Remove unused files (3713c4af) (@vinhnx)
- Update documentation for Phase 2.3 and Phase 3 completion; adjust status indicators and formatting (c14a4976) (@vinhnx)
- Enhance output filtering and logging configuration; add tests for ANSI background colors (9f6c659a) (@vinhnx)
- Remove production-ready implementation of VT Code Agent Skills (2876f86a) (@vinhnx)
- Implement enhanced file tracking and verification for skills execution (f8b0fe04) (@vinhnx)
- Enhance agent resilience by implementing anti-giving-up policies and constructive reasoning guidance (cf065688) (@vinhnx)
- Enhance skill execution and loading with new features and optimizations (9d0642f7) (@vinhnx)
- Refactor skill command exports and improve context manager imports (32e23054) (@vinhnx)
- Add dead code allowance for mentioned skills detection and loaded skills (c62cf652) (@vinhnx)
- Update tool policies to adjust permissions and enhance auto-injection documentation (232fc6fd) (@vinhnx)
- Refactor code executor documentation for clarity and consistency (cf6a9669) (@vinhnx)
- Implement automatic detection of Python interpreter for skill execution (8d55340e) (@vinhnx)
- Add automatic detection of Python interpreter for skill execution (1e63ab22) (@vinhnx)
- Enhance README and documentation with package manager preferences for skill execution (7fccb4e4) (@vinhnx)
- Fix formatting in README for package manager preferences (2489ee20) (@vinhnx)
- Add package manager preferences to skill rendering tests (d8622db2) (@vinhnx)
- Update Homebrew formula to version 0.49.7 (b92e872d) (@vinhnx)
- Update README with new badges and formatting (fd042397) (@vinhnx)
- Update README.md (d9932eba) (@vinhnx)
- Update links for Google Gemini and Devin DeepWiki (aa6806cb) (@vinhnx)
- Format cargo fmt (8dca0110) (@vinhnx)
- Update Homebrew formula to version 0.49.8 (8c7d2687) (@vinhnx)
- Update model constants and configurations; replace deprecated models and add new ones (bfba9cf6) (@vinhnx)
- Update Homebrew formula to version 0.50.1 (1915a078) (@vinhnx)
- Enhance skill manifests with additional metadata fields and validation rules (9fa622d8) (@vinhnx)
- Fix formatting in code review skill instructions (9e533847) (@vinhnx)
- Update Homebrew formula to version 0.50.2 (f7f28770) (@vinhnx)
- Update description of VT Code in README (867783e9) (@vinhnx)
- Update Homebrew formula to version 0.50.4 (8bb35457) (@vinhnx)
- Update Homebrew formula to version 0.50.5 (cd9efd28) (@vinhnx)
- Update Homebrew formula to version 0.50.7 (b4dde82d) (@vinhnx)
- Update tool policy mappings and switch default model to gpt-oss:120b-cloud in vtcode.toml (506b0fcd) (@vinhnx)
- Refactor tests, enhance path security, and implement rate limiting for tool execution (536dbbe8) (@vinhnx)
- Update Homebrew formula to version 0.50.9 (7500c402) (@vinhnx)
- Remove deprecated PDF processor and web search skills; update dependencies for workspace compatibility (fbc275c7) (@vinhnx)
- Add Hugging Face provider and integrate with existing LLM framework (5eb265fb) (@vinhnx)
- Downgrade anstyle-crossterm to 3.0 and crossterm to 0.28.1; refactor imports to use ratatui::crossterm; update related code for compatibility. (943f3059) (@vinhnx)
- Implement dual output execution for tools, enhancing token efficiency and backward compatibility; refactor UI widgets and session management for improved structure and readability. (5e51598f) (@vinhnx)
- Add force cancel PTY session functionality and update model constants (b2d25b81) (@vinhnx)
- Remove RL Optimization Loop details from README (29ec059b) (@vinhnx)
- Update default model and reorder tool policy entries. (9993af4d) (@vinhnx)
- Applied principles from CLIG.dev, POSIX, GNU, and Heroku CLI Style Guide. (da0db454) (@vinhnx)
- Add contributing, releasing, and security documentation; enhance doctor command with diagnostics and suggestions (35d23b87) (@vinhnx)
- Refactor references from "VTCode" to "VT Code" across documentation and source files (567841ec) (@vinhnx)
- Revert "feat: enhance input history navigation and improve session input handling" (36f217c6) (@vinhnx)
- Update README.md (47f47338) (@vinhnx)
- Refactor tool permission handling and improve terminal cleanup on exit; update default model and reasoning effort in configuration (cdfa093f) (@vinhnx)
- Implement memory optimizations and verification scripts (14efb508) (@vinhnx)
- Add memory monitoring and pressure detection system (0d05c101) (@vinhnx)
- Remove context trimming functionality and related configurations (cc1bcc24) (@vinhnx)
- Refactor code for improved readability and maintainability; update configuration files for minimal setup and enhanced security features. (3f68b2a6) (@vinhnx)
- Implement marketplace and plugin management system with configuration, installation, and registry features (6aa3c228) (@vinhnx)
- Implement tool permission handling in ask command and add Vim mode support in TUI (f7dfa4eb) (@vinhnx)
- Refactor status line configuration and add notification support across terminals (bab45872) (@vinhnx)
- Implement custom slash commands and enhance Vim mode functionality (a4943e59) (@vinhnx)
- Refactor custom slash command handling, enhance command validation, and improve error handling; update authoring and container documentation; add status check operation for code intelligence; implement plugin unloading; enhance tool registry with schema retrieval; improve project structure directory management. (d5be16ab) (@vinhnx)
- Refactor plugin manager and manifest handling; improve code readability and organization (f774f923) (@vinhnx)
- Add subagent system with specialized agents and spawn functionality (97ba9d4c) (@vinhnx)
- Refactor codebase for improved readability and functionality (fefe297d) (@vinhnx)
- Refactor codebase for improved readability and functionality (364178d8) (@vinhnx)
- Refactor documentation and comments to use "VT Code" consistently; remove timeline pane functionality and related configurations. (804522f2) (@vinhnx)
- Refactor turn processing result handling and improve parameter passing; clean up unused code and imports across various modules. (fbe30861) (@vinhnx)
- Add tool call/output pairing validation and normalization to conversation history (1546ea2c) (@vinhnx)
- Refactor test files and update configuration for agent provider (359fb5b3) (@vinhnx)
- Refactor tools to use unified search and file operations; remove deprecated methods and enhance error handling (a7bb59d3) (@vinhnx)
- Add methods to get tool by name and return workspace root as owned PathBuf (d8fe955d) (@vinhnx)
- Add Models Manager module with caching and model family definitions (81510903) (@vinhnx)
- Update tool policies and configuration for agent provider (206178d1) (@vinhnx)
- Implement shell command handler with sandbox support and timeout management (544147fd) (@vinhnx)
- Implement execution policy and sandboxing system with configurable permissions (a0853e16) (@vinhnx)
- Add skill packaging and validation scripts; update SKILL.md and installer scripts (abd6ab90) (@vinhnx)
- Bump system skills marker version to v2 for Codex compatibility updates (f64abf45) (@vinhnx)
- Refactor code for improved readability and consistency across handlers (4fb0628e) (@vinhnx)
- Add release monitoring and auto-install tools (40d4e0c8) (@vinhnx)
- Make GitHub release creation more robust with better error handling (bbe9e177) (@vinhnx)
- Update README with new documentation links (799f9a8a) (@vinhnx)
- Apply consistent formatting and whitespace adjustments (19c0d087) (@vinhnx)
- Fix linux release name (30d26c69) (@vinhnx)
- Optimize performance across various modules by reducing allocations, improving validation checks, and enhancing memory management strategies. (35e9f679) (@vinhnx)
- Refactor config loading and event handling for improved readability and efficiency (f9b3ce70) (@vinhnx)
- Refactor run_turn_handle_tool_failure and run_turn_handle_tool_cancelled to use parameter structs for improved readability and maintainability (958e4be7) (@vinhnx)
- Implement tool execution history, harness context, resiliency tracking, and timeout management (623ad414) (@vinhnx)
- Remove Vim mode implementation and related configurations (0ee7e606) (@vinhnx)
- Implement ACP Client V2 with JSON-RPC 2.0 support and session management (d0997e28) (@vinhnx)
- Integrate Phase 4 features: add circuit breaker, tool health tracker, and adaptive rate limiter; enhance validation caching; implement pattern detection in loop detector; refactor parallel execution; and improve tool execution metrics. (47e0028c) (@vinhnx)
- Add ask_user_question tool and enhance wizard modal functionality (328392c8) (@vinhnx)
- Implement Plan Mode with read-only exploration and user confirmation for execution (1b2c8e47) (@vinhnx)
- Add libc as an optional dependency on Linux (cbf79b20) (@gzsombor)
- Add new models to OpenRouter configuration and update ModelId enums (bc243ab4) (@vinhnx)
- Update tool policies to prompt for edit and unified search actions; add minimal configuration file generation (e19919f2) (@vinhnx)
- Add regenerate index command and update skills management (f9bf1dde) (@vinhnx)
- Enhance border styling based on editing mode and update trust mode handling in input widgets (f3693e22) (@vinhnx)
- Revert "Update tool policies to prompt for edit and unified search actions; add minimal configuration file generation" (91e9d8a0) (@vinhnx)
- Add OpenResponses streaming and types modules (b44fcdd2) (@vinhnx)
- Implement reasoning display toggle in chat UI and update related configurations (39fccfd4) (@vinhnx)
- Update logo image size in README.md (6a54ac9d) (@vinhnx)
- Refactor code for improved readability and consistency across multiple files (f6b83c3a) (@vinhnx)
- Implement diff preview feature for file changes approval (54e7face) (@vinhnx)
- Update apply_patch tool to use VT Code format and adjust related documentation (1b06aca5) (@vinhnx)
- Add circuit breaker functionality for error recovery and diagnostics (9fcdbf0c) (@vinhnx)
- Remove A2A CLI commands from README (5c13fac1) (@vinhnx)
- Refine wording in support section of README (82e27d14) (@vinhnx)
- Enhance token budget management and file operation indicators (53624e1b) (@vinhnx)
- Update TODO (96bdf091) (@vinhnx)
- Implement footer widget and sidebar for responsive layout; add todo detection and spinner updates (a28f781d) (@vinhnx)
- Revert "refactor: enhance history navigation and update inline event handling" (1787ef43) (@vinhnx)
- Revert "refactor: improve history navigation and update input handling" (73448a19) (@vinhnx)
- Update ACP implementation (53acb3c2) (@vinhnx)
- More refine ACP (fcf4253c) (@vinhnx)
- Remove logo and title section from README (d8888bac) (@vinhnx)
- Update README with title and badge links (d006999d) (@vinhnx)
- Refactor markdown rendering and improve event handling in TUI session (d1b11a77) (@vinhnx)
- More proper way to read spool files (a8912c2d) (@vinhnx)
- Update README (5a3c3648) (@vinhnx)
- Iimplement request_user_input tool, and update related components for structured user input. (a5872499) (@vinhnx)
- Fix build errors (26abed5c) (@vinhnx)
- Remove VT Code development support section (cece55da) (@vinhnx)
- Add adaptive logo (29e39639) (@vinhnx)
- Clarify VT Code as open-source coding agent (ab752ec2) (@vinhnx)
- Modify badge styles in README.md (a6f29e6c) (@vinhnx)
- Update TODO (eab58c20) (@vinhnx)
- Update TODO (eb09e382) (@vinhnx)
- Fix build error (5939b210) (@vinhnx)
- Update badge colors in README.md (e32a1e68) (@vinhnx)
- Fix build error (7d167ac2) (@vinhnx)
- Fix build error (ce058db4) (@vinhnx)
- Fix build docs.rs (d86f9de4) (@vinhnx)
- Fix badge links in README.md (90c605bc) (@vinhnx)
- Update Agent Skills badge in README (5bf6c889) (@vinhnx)
- Update README.md (f68b1f33) (@vinhnx)
- Fix build docs.rs (d6185275) (@vinhnx)
- Refactor badge links in README.md (5ec8128c) (@vinhnx)
- Update README with VT Code description (2b34f5f3) (@vinhnx)
- Update VHS script (2d9bf2b0) (@vinhnx)
- Change badge colors in README.md (16be2daf) (@vinhnx)
- Change badge colors in README.md (692edebf) (@vinhnx)
- Update TODO (b5c9f45b) (@vinhnx)
- Remove blank lines (da04cee6) (@vinhnx)
- Update loading (2d5aaea4) (@vinhnx)
- Add Kimi K2.5 Thinking to OpenRouter provider (7daad30f) (@vinhnx)
- Implement Open Responses specification with streaming events, output items, and response handling (81fb3a5c) (@vinhnx)
- Implement Open Responses integration and configuration options (43656ccc) (@vinhnx)
- Refactor code structure for improved readability and maintainability (df447d1c) (@vinhnx)
- Enhance Open Responses with sequenced events and improved item serialization (6244fd4f) (@vinhnx)
- Add technical whitepapers on security architecture and modular design principles (93d15f89) (@vinhnx)
- Enhance Open Responses specification conformance and update documentation (30259d42) (@vinhnx)
- Refactor Cross.toml to consolidate Docker configuration for cross-compilation (2c85e431) (@vinhnx)
- Fix docker build (db0e8c36) (@vinhnx)
- Fix vtcode-file-search build error (8437a13d) (@vinhnx)
- Improve deploy release (6b931dd7) (@vinhnx)
- Refactor tool outcome handling by consolidating success, failure, and timeout handlers, and updating tool repetition tracking to only count successful calls. (33a9b664) (@vinhnx)
- Refactor tool outcome handling in the agent runloop by introducing tool-specific retry limits, centralizing repetition tracking, and enhancing context conversion. (ce765244) (@vinhnx)
- Refactor tool call handling by centralizing execution, permission, and safety validation logic into dedicated outcome handlers and removing the execution module. (cfb8e1d7) (@vinhnx)
- Refactor session loop and tool outcomes; remove unused code and improve context handling (016ef738) (@vinhnx)
- Refactor code for improved readability and consistency across multiple files (fec995fc) (@vinhnx)
- Fix permission (e2511d7e) (@vinhnx)
- Update README.md to fix badge links (7b44cfa4) (@vinhnx)
- Implement OpenRouter OAuth PKCE authentication flow and related utilities (c338e631) (@vinhnx)
- Fix badge links in README.md (e098b997) (@vinhnx)
- Implement process group management and graceful termination for child processes (d7aac98b) (@vinhnx)
- Implement Agent Trace storage and specification for AI code attribution (d6748fc4) (@vinhnx)
- Enhance release.sh with direct GitHub binary upload via gh CLI (9168134d) (@vinhnx)
- Add automatic gh auth switch and scope refresh (1b5f38ec) (@vinhnx)
- Add CI trigger and comprehensive release flow guide (d071618c) (@vinhnx)
- Update CI (c9cf0a74) (@vinhnx)
- Update release (bb930a4e) (@vinhnx)
- Migrate LM Studio 0.4 REST API (4d12e993) (@vinhnx)
- Refactor utilities and enhance validation (47894d5d) (@vinhnx)
- Refactor HTTP client and image processing utilities; re-export from vtcode-commons for compatibility (d86b2399) (@vinhnx)
- Refactor code intelligence tool to remove LSP dependencies and update documentation (83c90149) (@vinhnx)
- Refactor code for improved readability and consistency across multiple provider files; update deprecated configurations in vtcode.toml. (9234961d) (@vinhnx)
- Update format (20896f23) (@vinhnx)
- Fix syntax error on release yml (e3f3c98e) (@vinhnx)
- Fix warnings (b579708e) (@vinhnx)
- Clean (35a77d1a) (@vinhnx)
- Fix CI (d64036d3) (@vinhnx)
- Add star history section to README (3de92245) (@vinhnx)
- Duplicate badge links for Agent Skills and protocols (b9216314) (@vinhnx)
- Fix release note (f7562c14) (@vinhnx)
- Update model references from gpt-4 to gpt-5 across documentation and codebase (cec3d7c9) (@vinhnx)
- Update model references to "claude-haiku-4-5" across configuration and tests (b06501f4) (@vinhnx)
- Refactor agent guidelines, improve spacing in TODO, and enhance model picker logic; update Anthropic provider tests and configuration (0dece6ac) (@vinhnx)
- Revert "refactor: enhance base URL resolution and improve JSON handling in request builder" (4cb8f2d9) (@vinhnx)
- Improve release (ff4ae644) (@vinhnx)
- Update skill paths to use .agents/skills and deprecate .vtcode/skills (e20b2474) (@vinhnx)
- Add webapp-testing skill with Playwright scripts and examples; introduce xlsx skill for spreadsheet handling (e2dfd86a) (@vinhnx)
- Update TODO (fb065df0) (@vinhnx)
- Refactor diff handling and rendering for improved clarity and summary display (f8dbf9e1) (@vinhnx)
- Implement no_spool functionality for tool output and enhance cursor behavior during scrolling and shimmer states (477209fd) (@vinhnx)
- Refactor MCP integration and update dependencies (b31b5407) (@vinhnx)
- Improve CI CD (79589790) (@vinhnx)
- Update script (9b0ce937) (@vinhnx)
- Update script (77c4320b) (@vinhnx)
- Update script (6b172935) (@vinhnx)
- Implement command caching and gatekeeper policy; enhance file reading with async logging and performance tracking (e464ee31) (@vinhnx)
- Refactor test assertion for compact_title method in MCP event (4f44b68a) (@vinhnx)
- Refactor spinner implementation and enhance cursor behavior during status updates (0b6828ac) (@vinhnx)
- Cleanup TODO.md by removing outdated tasks and enhancing UI transition notes (21d1d5c6) (@vinhnx)
- Enhance input widget styling with padding and background; update configuration theme and editing mode (958f7c38) (@vinhnx)
- Add skills-ref commands for skill validation, listing, and prompt generation; update skill discovery paths and manifest structure (df2919d9) (@vinhnx)
- Implement tool safety checks, enhance wizard modal, and update configuration (574d60de) (@vinhnx)
- Add support for Claude Opus 4.6 model with adaptive thinking and update related configurations (77d0d485) (@vinhnx)
- Add context management support to LLM requests and related components (e893cee1) (@vinhnx)
- Add adaptive thinking support for Claude Opus 4.6 model and update related configurations (19515dee) (@vinhnx)
- Add support for effort parameter in Anthropic API and related validation (9519a720) (@vinhnx)
- Add jq dependency check and improve cargo release process (5c9a2c82) (@vinhnx)
- Remove code repetition and special casing of local providers (ec4b2099) (@gzsombor)
- Update script (35c8d01f) (@vinhnx)
- Add experimental agent teams feature with slash commands and configuration (b70ce06e) (@vinhnx)
- Improve output spooler and system prompt handling (255e0a44) (@vinhnx)
- Enhance user input tools to restrict usage to Plan mode only (93675177) (@vinhnx)
- Refactor CLI argument documentation for clarity and consistency (087be239) (@vinhnx)
- Implement collapsible pasted message handling and improve image path parsing (5efac36e) (@vinhnx)
- Add team context and teammate management features (b322e6cd) (@vinhnx)
- Immprove (e3fc8d93) (@vinhnx)
- Revert "feat: integrate `tui_input` crate for enhanced input management and modal search functionality" (b6e27465) (@vinhnx)
- Refactor to use rustc_hash::FxHashMap for improved performance and memory efficiency; update related structures and configurations. (3b166144) (@vinhnx)
- Remove outdated optimization notes and focus on DRY opportunities in the codebase (c9ac418b) (@vinhnx)
- Refactor optimizer and tool result handling; enhance turn duration recording; update validation and state management; optimize loop detection; improve LLM request handling; remove fallback chains module; streamline tool execution checks; fix TUI modal search handling; adjust integration tests for tool usage. (9df107c3) (@vinhnx)
- Add safety validation and transition functions for plan mode handling (9db46b25) (@vinhnx)
- Add prompt assembly mode and enhance tool validation (480ed33a) (@vinhnx)
- Enhance plan mode handling and tool safety validation; refactor prompt management and session loop logic (63a40249) (@vinhnx)
- Implement tool catalog state management and integrate into MCP tool lifecycle (18c73b54) (@vinhnx)
- Add GLM-5 model support and remove deprecated GLM-4.5/4.6 models (29d0992a) (@vinhnx)
- Update models and configurations for Gemini 3 and GLM-5; adjust tool capabilities and user confirmations (7f05b778) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Update TODO.md (3391d6b9) (@vinhnx)
- Add spool chunk read tracking and enforce limits per turn (c09ef6d3) (@vinhnx)
- Increase max tool calls per turn to 48 and implement budget warning system in harness (7b6cade0) (@vinhnx)
- Rename UpdatePlanCommand to TaskTrackerCommand and refactor related files and documentation (b4520273) (@vinhnx)
- Add plan-mode task tracker with CRUD functionality and integrate with existing tools (56c7e3b8) (@vinhnx)
- Add navigation loop guidance and improve plan mode handling (5d162ae6) (@vinhnx)
- Refactor user input option generation and enhance markdown rendering in modals (a72a63c0) (@vinhnx)
- Add reference to git-cliff for changelog generation (8dd703f1) (@vinhnx)
- Refactor tool call handling and configuration limits (fcc783dd) (@vinhnx)
- Refactor registry and facade methods for improved error handling and readability; update configuration defaults and themes. (5ea7b7e1) (@vinhnx)
- Refactor error handling in various modules to improve resilience and logging; enhance cache management and concurrency controls. (035a14bb) (@vinhnx)
- Enhance TTY integration using crossterm's IsTty for consistent terminal handling (2a4e71d6) (@vinhnx)
- Implement unified exec terminal panel and enhance PTY session handling (f8505ba9) (@vinhnx)
- Refactor run command output handling and improve command summary rendering (59e8fab8) (@vinhnx)
- Implement plan artifact write handling and adjust LLM timeout logic (5edb5d66) (@vinhnx)
- Enhance timeout messages to include "continue" nudge for stalled turns in plan mode (03cd0a65) (@vinhnx)
- Enhance plan mode interview options with recommended choices for goals, constraints, and verification (3263981c) (@vinhnx)
- Refactor tool interaction: replace `ask_user_question` with `request_user_input`, enhance error classification, and improve function declaration caching. (2af17e60) (@vinhnx)
- Remove unified symbol extractor and related patterns; update language summarization to use file extensions; enhance README for clarity and feature descriptions; clean up unused imports in tools module. (aca4dbef) (@vinhnx)
- Implement XDG Base Directory Specification, add Zed extension file search commands, enhance PTY ANSI handling, and improve PTY pipe infrastructure. (03705867) (@vinhnx)
- Update documentation links and paths across multiple files (76d81123) (@vinhnx)
- Update LLM provider to Gemini and change default model and theme (9b20f88e) (@vinhnx)
- Fix gemini streaming processor and update TODO.md (04927325) (@vinhnx)
- Refactor model definitions and remove deprecated GPT-5.1 variants; introduce GPT-5.3 Codex with updated descriptions and presets. (6851ba32) (@vinhnx)
- Update badge links in README.md (b74377e3) (@vinhnx)
- Refactor code for improved readability and performance; simplify conditionals and remove unnecessary clones (93989d58) (@vinhnx)
- Add notification system with configurable preferences and delivery modes (e3be6b6b) (@vinhnx)
- Remove deprecated tool aliases and clean up related code for user input handling (5a23ab29) (@vinhnx)
- Unify ModelId and Provider across vtcode-config and vtcode-core; add regression tests to prevent future drift (3bef6a4d) (@vinhnx)
- Remove deprecated model ID and provider configurations, streamline model handling and update references across the codebase. (ebd573ec) (@vinhnx)
- Improve (d8ec16e4) (@vinhnx)
- Refactor code for improved readability and maintainability; streamline model handling and enhance diff preview functionality. (58607c7b) (@vinhnx)
- Add Qwen3.5-397B-A17B model support and update related identifiers and descriptions (a9e62f9c) (@vinhnx)
- Add turn elapsed timer feature and configuration options (f6867385) (@vinhnx)
- Remove xlsx skill and related files; update turn outcome message format (c31af51e) (@vinhnx)
- Implement exact token counting for various LLM providers and enable token counting feature (136c1b1f) (@vinhnx)
- Remove unused question-related policies and enhance token usage logic in ContextManager (4e39398a) (@vinhnx)
- Remove xAI Grok models and related configurations from the codebase (7b069620) (@vinhnx)
- Enhance file reference extraction to handle npm scoped packages and improve path validation (00646152) (@vinhnx)
- Enhance UI styling for PTY output and improve color handling across components (36cc1ca0) (@vinhnx)
- Add sidebar, slash command palette, and transcript widgets; implement session options and host adapter traits (fdc70374) (@vinhnx)
- Add utilities for ANSI parsing, caching styles, and diff generation (a20742ed) (@vinhnx)
- Update configuration for agent provider and default model; adjust theme settings (465ed6e7) (@vinhnx)
- Enhance changelog generation with Highlights and Contributors sections (8e25479e) (@vinhnx)
- Enhance tool output handling and caching logic (8635f4a0) (@vinhnx)
- Refactor unified_exec tool to add inspect action and improve documentation (eac81276) (@vinhnx)
- Refactor guidance messages for clarity and conciseness; update tool output handling and error messages. (ae454f4b) (@vinhnx)
- Add 'continue' action to unified_exec tool; refactor related logic and documentation (b2102ff4) (@vinhnx)
- Refactor tool policies and enhance shell command parsing; add next_continue_args handling (327a2467) (@vinhnx)
- Add clear screen and copy latest assistant reply commands; update related documentation and tests (39b2d94a) (@vinhnx)
- Enhance command detection by normalizing natural language requests and adding support for various command patterns (bc2be0b9) (@vinhnx)
- Change terminal theme from 'ciapre-dark' to 'mono' (6b03347c) (@vinhnx)
- Add appearance configuration and improve modal handling (afd21755) (@vinhnx)
- Enhance external editor integration and rewind functionality (e5f722ce) (@vinhnx)
- Refactor code for improved readability and functionality; normalize unified search arguments; enhance validation error handling; update default model in configuration. (56adb18f) (@vinhnx)
- Refactor palette handling and LLM retry logic; improve error handling and tests (64f0f949) (@vinhnx)
- Enhance task tracker functionality and error handling (42781438) (@vinhnx)
- Remove XAI provider tests and related assertions from model validation (880ec280) (@vinhnx)
- Remove deprecated integration and workflow tests; clean up test module imports (3c9f5f42) (@vinhnx)
- Update model references from "gemini-3-pro-preview" to "gemini-3.1-pro-preview" across the codebase (468a0765) (@vinhnx)
- Refactor caching mechanisms to use Mutex instead of RwLock for improved concurrency (af7ebf32) (@vinhnx)
- Refactor sandbox configuration and enhance ANSI capabilities (cfb2f51a) (@vinhnx)
- Update default model to GPT-5.3 Codex and enhance terminal title management (9d3d7d38) (@vinhnx)
- Update default model to GPT-5.3 Codex, change provider to OpenAI, and enhance reasoning effort configuration (24d3af83) (@vinhnx)
- Remove xAI provider support and update default models in configuration files (ad6290aa) (@vinhnx)
- Refactor code for improved readability and consistency in function calls (91fdb43a) (@vinhnx)
- Update code structure for improved readability and maintainability (24d81c35) (@vinhnx)
- Refactor themes and improve UI handling (970dadf8) (@vinhnx)
- Add support for native web search tools in Anthropic API (84c50dfb) (@vinhnx)
- Add support for file content parts in message handling and OpenAI provider (4effd39c) (@vinhnx)
- Update theme to "ciapre" across configurations and add new themes to the theme registry (9c8d0c16) (@vinhnx)
- Refactor syntax highlighting implementation and improve theme handling (a9236cfc) (@vinhnx)
- Enhance tool routing and UI theme accessibility (0747f3d8) (@vinhnx)
- Enhance diff rendering and suppress redundant recap in TUI (0f3513b1) (@vinhnx)
- Fix light theme rendering issues and improve accessibility; enhance theme handling and live reload functionality (371a681b) (@vinhnx)
- Normalize unified search arguments and enhance argument extraction logic (a732a449) (@vinhnx)
- Fix light theme rendering issues and improve theme handling for better accessibility (ec131237) (@vinhnx)
- Refactor diff rendering and styling for consistency and accessibility (3edcf204) (@vinhnx)
- Refactor token counting methods and remove unused tests across providers; update configuration for default provider and model. (55c76984) (@vinhnx)
- Add health check and reconnection methods to MCP provider; enhance error classification and guidance (45fcc200) (@vinhnx)
- Enhance error handling and logging across providers; improve session archive retention and diagnostics (ce94ca83) (@vinhnx)
- Refactor error recovery to use tokio::sync::RwLock; update tool outcome handling and improve cache retrieval methods (76c1789f) (@vinhnx)
- Add support for "show diff" command and improve tool call validation (62051691) (@vinhnx)
- Update color scheme for unified diff background and gutter colors; change terminal theme to improve readability (525653e9) (@vinhnx)
- Add diff theme configuration and improve styling for terminal output (501634f1) (@vinhnx)
- Enhance interactive mode with new keyboard shortcuts and shell mode indicators; update configuration to disable turn timer (87919849) (@vinhnx)
- Refactor input handling and padding logic; enhance shell command validation and error feedback (d520d2ea) (@vinhnx)
- Add file reference metadata handling and update input with aliases (91e39103) (@vinhnx)
- Implement postamble exit summarization with metric calculation and enhance markdown output styling (02d01e99) (@vinhnx)
- Implement code change tracking and exit summary reporting (5e776910) (@vinhnx)
- Refactor session handling to streamline resume logic and enhance message rendering (6654f328) (@vinhnx)
- Enhance skill authoring and routing logic; add response chaining support (121a8ece) (@vinhnx)
- Add `thiserror` integration for improved error handling across multiple modules (b5081d89) (@vinhnx)
- Add websocket connection limit error handling and update API versioning (6f76cfb7) (@vinhnx)
- Add installation scripts to release artifacts and update installation documentation (816c6d41) (@vinhnx)
- Refactor error log collector to use a bounded buffer and enhance link rendering in markdown (c7ef605b) (@vinhnx)
- Refactor line truncation logic and introduce utility module for consistent ellipsis handling (69742626) (@vinhnx)
- Refactor websocket header insertion and update apply_patch handler to use Arc for session and turn context (13e4b403) (@vinhnx)
- Add error log buffer size limit and refactor tool context for constrained policies (f8b71916) (@vinhnx)
- Refactor tool routing and permission handling; improve URL-aware text wrapping (564fdad7) (@vinhnx)
- Refactor tool permission caching and enhance error handling; improve text wrapping utilities (ed68b7fd) (@vinhnx)
- Add support for zsh exec wrapper and enhance PTY configuration (9d992fed) (@vinhnx)
- Add data-oriented design plan and optimization phases for vtcode (98175726) (@vinhnx)
- Enhance status line configuration and improve token extraction logic for file references (68bb39d4) (@vinhnx)
- Document applied default status line items pattern and outline data-oriented design plan for vtcode (db98b328) (@vinhnx)
- Add tests for PtyConfig validation and refactor session creation methods (8cd350ef) (@vinhnx)
- Refactor TODO.md to improve formatting and clarify implementation instructions (265c835a) (@vinhnx)
- Enhance file palette key handling to manage selection and normal submit behavior (309e8b22) (@vinhnx)
- Refactor process group termination logic for robust child process cleanup (bb1ab07f) (@vinhnx)
- Refactor WebSocket session handling and enhance incremental input validation (e874c269) (@vinhnx)
- Refactor sandbox permissions and enhance tool execution handling (f997edb2) (@vinhnx)
- Implement fine-grained rejection controls for approval prompts and update related logic (3b9a7775) (@vinhnx)
- Refactor caching mechanisms to use inner struct with RwLock for improved concurrency (46102de4) (@vinhnx)
- Refactor session initialization to encapsulate execution and metadata contexts (ef3d1020) (@vinhnx)
- Enhance TurnProcessingContext with new as_turn_processing_context method for improved context management (c0b47e4d) (@vinhnx)
- Refactor RunLoopContext and TurnProcessingContext to use constructor methods for improved clarity and maintainability (cd9d63a5) (@vinhnx)
- Refactor TurnProcessingContext to use parts for better modularity and readability (b133ff3d) (@vinhnx)
- Refactor code for improved readability and consistency in multiple files (2565bbdb) (@vinhnx)
- Add Clippy configuration for vtcode with custom lint settings (fac8a298) (@vinhnx)
- Implement theme synchronization for Alacritty, Ghostty, and Kitty terminals; remove CachedStyleParser and replace with LrMap for improved performance; add terminal color probe for enhanced theme detection; update Cargo.toml files for Rust 1.85 compatibility. (c7dd0f28) (@vinhnx)
- Refactor code to use `parking_lot::RwLock` instead of `std::sync::RwLock` for improved performance; update various imports and method calls across multiple files; bump Rust version to 1.88 in Cargo.toml files. (10892d08) (@vinhnx)
- Refactor code for improved readability and consistency across multiple files (d876e596) (@vinhnx)
- Refactor command processing, improve output handling, and enhance telemetry reporting (ec20b546) (@vinhnx)
- Refactor tests and improve code readability (2c2ade2b) (@vinhnx)
- Add color policy module and integrate into styling and error display (3287a044) (@vinhnx)
- Add fuzzing adoption plan for security parsers and initial setup (4b4d8d8f) (@vinhnx)
- Refactor code for consistency and readability across multiple modules (502cda47) (@vinhnx)
- Refactor middleware chain initialization to use push method (2368dec1) (@vinhnx)
- Update integration tests to replace 'cargo test' with 'cargo check' and adjust access count assertions (862c07df) (@vinhnx)
- Add fuzzing support for security parsers and update documentation (bd445376) (@vinhnx)
- Refactor command safety checks and improve loop detection tests (b8ede10b) (@vinhnx)
- Enhance command validation and tool execution with unified action handling (0c3d4b86) (@vinhnx)
- Update script (d537fa69) (@vinhnx)
- Format code for readability and consistency (b14abfa6) (@vinhnx)
- Update terminal theme to ciapre-blue (80083b1a) (@vinhnx)
- Update code structure for improved readability and maintainability (049b0fcb) (@vinhnx)
- Add conversation compaction feature for Responses providers (2cb46d21) (@vinhnx)
- Implement adaptive task tracker with Plan Mode support and enhance checklist persistence (4d4ad68e) (@vinhnx)
- Remove spawn_subagent tool and related references (819fd210) (@vinhnx)
- Refactor agent state management and remove delegate mode references (a6f7ce4a) (@vinhnx)
- Refactor skill and MCP management for on-demand activation; improve dynamic context initialization and logging (22784ca6) (@vinhnx)
- Refactor session archive management; add debug log path handling and improve session identifier reservation (a3f71a41) (@vinhnx)
- Refactor tool policies; add support for cargo and git commands, and remove MCP initialization status check from session UI (4afee7a3) (@vinhnx)
- Refactor system prompt structure; enforce inclusion of references and next action questions in multi-file summaries (35508373) (@vinhnx)
- Refactor tools to unify search functionality and remove deprecated grep_file and list_files aliases (0742cf3a) (@vinhnx)
- Refactor tool constants; remove legacy aliases and add new tool IDs for improved clarity (0016a628) (@vinhnx)
- Refactor tool message handling and improve plan mode instructions (915627de) (@vinhnx)
- Enhance loop detection logic and improve interview questions for plan mode (25ad8380) (@vinhnx)
- Enhance loop and circuit breaker failure detection logic; improve agent's autonomous recovery capabilities (e4134225) (@vinhnx)
- Refactor turn processing to include reasoning details; update configuration for Minimax provider (fd61b5bd) (@vinhnx)
- Refactor slash area handling and update default model and editing mode in configuration (272b036d) (@vinhnx)
- Refactor widgets: remove modal, palette, and slash; update sidebar and session handling (2918aec6) (@vinhnx)
- Enhance system prompt guidelines: add principles for consistency, simplicity, and uncertainty recognition (ff8d53df) (@vinhnx)
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59) (@vinhnx)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998) (@vinhnx)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0) (@vinhnx)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d) (@vinhnx)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58) (@vinhnx)
- Implement interactive settings manager and refactor related components (1f2d7f2a) (@vinhnx)
- Update code structure for improved readability and maintainability (2ea69033) (@vinhnx)
- Update TODO (86150ea2) (@vinhnx)
- Update TODO (c0e49629) (@vinhnx)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688) (@vinhnx)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf) (@vinhnx)
- Standardize error handling policies and improve retry logic across components (e4c248c9) (@vinhnx)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885) (@vinhnx)
- Implement interactive skills management and enhance plan mode functionality (a0d19bc4) (@vinhnx)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681) (@vinhnx)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7) (@vinhnx)
- Update TODO (45cc3a6a) (@vinhnx)
- Enhance interactive skills management with new commands and aliases (471cd578) (@vinhnx)
- Refactor slash commands and improve skills management interface (f0391983) (@vinhnx)
- Update TODO (c7c11e12) (@vinhnx)
- Update TODO (f03eb648) (@vinhnx)
- Enhance Plan Mode interview handling and UI adjustments (77b8c823) (@vinhnx)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509) (@vinhnx)
- Refactor error handling and improve code safety across multiple modules (74ee2dea) (@vinhnx)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370) (@vinhnx)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f) (@vinhnx)
- Add native plugin support and enhance skill loading (8ba9fa9e) (@vinhnx)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b) (@vinhnx)
- Add native plugin system implementation and documentation (0c0c856b) (@vinhnx)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670) (@vinhnx)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc) (@vinhnx)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f) (@vinhnx)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7) (@vinhnx)
- Add model_search_value function for generating model search strings (7ba202a8) (@vinhnx)
- Remove production.py and version.rs files from the project (f30fa2e2) (@vinhnx)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300) (@vinhnx)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3) (@vinhnx)
- Refactor code structure for improved readability and maintainability (43dd252e) (@vinhnx)
- Refactor imports and update theme in configuration (a9517434) (@vinhnx)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68) (@vinhnx)
- Add memchr dependency and update usage in ANSI parser (1b16a714) (@vinhnx)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e) (@vinhnx)
- Refactor imports and update configuration settings across multiple files (05d95e7b) (@vinhnx)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b) (@vinhnx)
- Fix path (cc0eaa6a) (@vinhnx)
- Update API key and model provider in vtcode.toml (2939fceb) (@vinhnx)
- Implement task guards for background operations and improve async handling in various modules (23a114ab) (@vinhnx)
- Implement interactive diagnostics with quick and full options in the slash command (6876ae63) (@vinhnx)
- Implement Markdown parsing with support for lists, links, and tables (0ded92d5) (@vinhnx)
- Implement prompt cache shaping mode and enhance caching strategies (5f43940d) (@vinhnx)
- Implement history normalization for LLM requests and enhance execution context management (ad977a32) (@vinhnx)
- Refactor tool output helpers and improve cache management (3c825982) (@vinhnx)
- Refactor configuration layer loading and error handling; add support for tracking layer origins in merges; implement API for reading and writing configuration with version checks. (5fe3d2da) (@vinhnx)
- Refactor shell command building, update justification messages, enhance search metrics, improve web fetch domain matching, and adjust tool policies in configuration. (3c5d7600) (@vinhnx)
- Refactor code for improved readability and consistency across multiple files (adb2ad30) (@vinhnx)
- Refactor Cargo.toml files to use workspace dependencies and update versioning scheme (a7bf7eea) (@vinhnx)
- Add MCP requirements enforcement and update related configurations (1f39beec) (@vinhnx)
- Update OpenAI model references to GPT-5.4 and adjust reasoning effort defaults (38c7644a) (@vinhnx)
- Implement shell command approval enhancements and justification handling (3ff40170) (@vinhnx)
- Refactor session mode handling and update related components (d78397ef) (@vinhnx)
- Update TODO (ae3651f1) (@vinhnx)
- Update TODO (fb5ee65f) (@vinhnx)
- Refactor modal handling to use overlay system and improve session state management (d6171cfc) (@vinhnx)
- Enhance exec command functionality with session resumption and prompt handling (d234fee7) (@vinhnx)
- Add review command and related functionality (ef0298c7) (@vinhnx)
- Refactor tool execution and safety gateway: remove legacy orchestrator and unified executor, streamline parallel execution, and enhance safety context handling. (2946fcf7) (@vinhnx)
- Add support for resuming and forking sessions across workspaces (32bd498f) (@vinhnx)
- Add structured error handling with VtCodeError and ErrorCode (93beb517) (@vinhnx)
- Consolidate reliability plan (47186a92) (@vinhnx)
- Refactor ToolRegistration to use Arc<str> for name and add unregister_tool method in ToolRegistry (cba4392f) (@vinhnx)
- Refactor tool policy handling and enhance safety checks for file access (2e9a9c16) (@vinhnx)
- Refactor command session handling and rename PTY references to unified_exec (80023410) (@vinhnx)
- Refactor tool execution handling to unify command names and improve compatibility (e1f3b602) (@vinhnx)
- Refactor tool intent handling to unify command names and improve output processing (cb0581be) (@vinhnx)
- Refactor tool execution handling and update command aliases to use unified_exec (097b5d85) (@vinhnx)
- Add support for creating and managing PTY sessions in the tool registry (8cbfcb87) (@vinhnx)
- Add support for optional patch input in apply_patch handler and update related schemas (29ed7b6e) (@vinhnx)
- Add assistant phase handling to message processing and tests (f448df1c) (@vinhnx)
- Add support for OpenAI service tier configuration and handling (5586bffc) (@vinhnx)
- Implement VT Code ast-grep installer (f8901da1) (@vinhnx)
- Fix build release (f9f272e9) (@vinhnx)
- Refactor pattern detection and command safety checks for improved efficiency and optimize tool name caching. (4abdbc6f) (@vinhnx)
- Review ZedAgent registry mutex (3ead1b8b) (@vinhnx)
- Apply async docs updates (3b75f5cf) (@vinhnx)
- Incorporate async docs guidance (800f1927) (@vinhnx)
- Enhance task listing filters (24b35cee) (@vinhnx)
- Improve batched request handling (e1cbedf5) (@vinhnx)
- Improve ANSI styles (cbe58ce9) (@vinhnx)
- Refactor agent loop to support initial configuration and enhance live reload functionality (7aaba845) (@vinhnx)
- Review vtcode PR 13090 updates (e4ee2358) (@vinhnx)
- Refactor SandboxPolicy to support network access and allowlisting in ReadOnly mode (89127618) (@vinhnx)
- Enhance Anthropic tool support with code execution and memory tools; improve validation and error handling (fac76a15) (@vinhnx)
- Implement referenced Rust PR updates (c43428b1) (@vinhnx)
- Fix @file diff lookup (183a1ec3) (@vinhnx)
- Refactor tool routing and ACP (c37388a6) (@vinhnx)
- Add headings to vtcode configs (b0f294f7) (@vinhnx)
- Remove forced budget stop logic (6ec74d5c) (@vinhnx)
- Enhance interleaved thinking model support and serialization across providers (a8e8bd27) (@vinhnx)
- Refactor task tracker rendering and state management (06566014) (@vinhnx)
- Implement workspace environment variable management and interactive selection UI (7e94f5df) (@vinhnx)
- Add local tool execution and session management for ZedAgent (1a460fc2) (@vinhnx)
- Refactor session setup and tool execution: remove unused code and improve clarity (06cc7671) (@vinhnx)
- Implement ast-grep installer with caching and locking mechanisms; add ripgrep installer with platform-specific installation logic; enhance debug logging capabilities. (fd0dadc9) (@vinhnx)
- Clean unused rust code (98f12364) (@vinhnx)
- Refactor agent configuration: remove unused code, enhance API key environment handling, and streamline model selection logic (89cbd68a) (@vinhnx)
- Add Gemini wire models and streaming processor implementation (4114e15c) (@vinhnx)
- Refactor approval policy handling: streamline rejection logic and enhance clarity in related modules (10630d73) (@vinhnx)
- Refactor sandbox policy handling and improve test coverage (8bfd4efa) (@vinhnx)
- Refactor error handling and add fallback mechanisms for model not found scenarios in OpenAI and OpenResponses providers (12f7a717) (@vinhnx)
- Refactor provider configuration: add timeout and specific OpenAI/Anthropic options (288cb188) (@vinhnx)
- Implement file conflict monitoring and auditing features (07e4b493) (@vinhnx)
- Refactor code for improved readability and consistency across file operations and conflict detection (6354965c) (@vinhnx)
- Refactor file conflict handling and introduce runtime execution management (e60faf6a) (@vinhnx)
- Refactor update handling and relaunch logic; enhance update command execution strategy (dffb89da) (@vinhnx)
- Enhance tool policies and follow-up prompt detection; update theme configuration (1463e225) (@vinhnx)
- Update notes (aefceac2) (@vinhnx)
- Enhance TurnLoopContext and processing logic for improved tool call handling and metadata management (327379f2) (@vinhnx)
- Add exhausted_tool_call_limit method and update budget handling logic (ebdc1b6f) (@vinhnx)
- Add LiteLLM provider support and documentation (327183cb) (@vinhnx)
- Enhance parameter sanitization and alias handling for tool arguments (190ee131) (@vinhnx)
- Update TODO (3b2e19fc) (@vinhnx)
- Add search tools status badge and enhance language detection (7150057d) (@vinhnx)
- Fix install script (25a0e060) (@vinhnx)
- Update badge colors and links in README.md (99175a26) (@vinhnx)
- Update README.md (b433aefb) (@vinhnx)
- Update TODO (163e9716) (@vinhnx)
- Refactor API key handling and update model reasoning capabilities (db0bdf03) (@vinhnx)
- Refactor reasoning handling in OpenRouterProvider and improve model checks (f237aa93) (@vinhnx)
- Refactor configuration loading to use workspace paths and remove deprecated environment variable handling (352cc9e5) (@vinhnx)
- Remove outdated model picker improvement notes from TODO (a345f1e4) (@vinhnx)
- Overhaul model picker UI (b0f33444) (@vinhnx)
- Enhance prompt caching and telemetry metrics; update configurations for improved cache handling (3fb970ac) (@vinhnx)
- Update TODO (3ee104de) (@vinhnx)
- Implement task metadata handling and continuation logic in task tracker (96b8362a) (@vinhnx)
- Add effective_exec_events_path function and related tests; enhance harness_exec output normalization (05d86d58) (@vinhnx)
- Update notes (0664f0b9) (@vinhnx)
- Refactor tool output handling to remove follow-up prompts and streamline continuation logic (bd4000aa) (@vinhnx)
- Update note (52cdad90) (@vinhnx)
- Refactor execution result handling to improve metadata management and streamline output processing (a9ada48e) (@vinhnx)
- Enhance loop recovery hints and streamline error handling in tool execution (fabb03ed) (@vinhnx)
- Refactor tool output rendering to improve markdown handling and add hidden lines notice (62f7ad22) (@vinhnx)
- Refactor tool output rendering to streamline follow-up hints and improve output handling (51e754ac) (@vinhnx)
- Add support for Nemotron-3-Super model and improve terminal file path handling (33f04db1) (@vinhnx)
- Add exact terms match function and update default model to Nemotron-3-Super (8df4f93e) (@vinhnx)
- Add tests for file path handling and improve location suffix resolution (71638904) (@vinhnx)
- Add reference link for control sequences in TODO.md (d951810e) (@vinhnx)
- Remove empty code change entries from the changes log (fb89f6c2) (@vinhnx)
- Add support for tool-free recovery mode and enhance wait feedback mechanisms (8c9f4ca1) (@vinhnx)
- Refactor exit summary display for single and multi-model sessions; improve Ctrl+C signal handling for immediate exit (45558686) (@vinhnx)
- Enhance chat request parsing to preserve assistant phase and streamline message handling (cc3a923b) (@vinhnx)
- Enhance performance by optimizing vector initializations and memory allocations across multiple files (239ac831) (@vinhnx)
- Add 'tty' parameter for PTY execution and enhance output handling in exec responses (049c4382) (@vinhnx)
- Enhance error handling and output guidance for command execution failures (f1cdd4f2) (@vinhnx)
- Enhance error handling and recovery guidance for tool execution failures (1f3445c0) (@vinhnx)
- Refactor output handling to replace read_file and grep_file with unified_file and unified_search (5e094ba7) (@vinhnx)
- Replace read_file and grep_file references with unified_file and unified_search across multiple files (d9794cf4) (@vinhnx)
- Refactor unified_exec argument validation and enhance preflight checks (3d5c56c1) (@vinhnx)
- Refactor loop detection and recovery policies for unified execution (bb764846) (@vinhnx)
- Update continuation policy to default to "all" and implement pause/resume functionality in exec mode (c7b98951) (@vinhnx)
- Add default harness log directory and update event log path documentation (3135c322) (@vinhnx)
- Refactor input handling: update keybindings and queue submission logic (598a2ad5) (@vinhnx)
- Implement session IDE toggle (28c6a858) (@vinhnx)
- Add ANSI escape sequences for scroll region and insert/delete operations (339b6053) (@vinhnx)
- Update TODO (fe850c3d) (@vinhnx)
- Enhance execution flow: add non-interactive settling logic and improve test coverage (f7047954) (@vinhnx)
- Update TODO (45a02148) (@vinhnx)
- Add support for tracking and reusing successful readonly tool signatures (e97063ad) (@vinhnx)
- Add tool_call_id to event items and enhance usage tracking (7ab2319c) (@vinhnx)
- Add tool_call_correlation_id function for fallback handling in output items (134b5691) (@vinhnx)
- Enhance command detection: support likely typos and improve tool call correlation handling (2b64d276) (@vinhnx)
- Refactor ANSI handling: centralize constants, improve documentation, and enhance utility functions (d268d5c9) (@vinhnx)
- Implement auto-compaction for turn loop and enhance skill templates with routing guidance (31aef8a1) (@vinhnx)
- Refactor plan file template for clarity and structure (df3764eb) (@vinhnx)
- Add support for blocked handoff events and quiet success output in hooks (f978598b) (@vinhnx)
- Update AGENTS instructions (b584d382) (@vinhnx)
- Add fallback filenames to ProjectDocOptions and related functions; enhance clickable link handling in ANSI renderer; implement session history persistence settings; introduce new FileOpener and HistoryPersistence enums in config. (3e4f1ea1) (@vinhnx)
- Add request_permissions field to RejectConfig and update related logic (54430912) (@vinhnx)
- Add support for explicit transcript links in the TUI session (31add726) (@vinhnx)
- Enhance exit summary output with model and reasoning details; update configuration for new provider and model (228f5909) (@vinhnx)
- Implement CGP integration facade for ToolRegistry with pipeline support (68bd7f63) (@vinhnx)
- Add direct tool completion handling and refactor related logic (ab246cb8) (@vinhnx)
- Scan commits for potential bugs (c6b7ac99) (@vinhnx)
- Implement mouse selection and interaction for modal lists and input areas (e618c0da) (@vinhnx)
- Enhance observability and model validation for LLM requests and tool executions (fc5c883a) (@vinhnx)
- Update documentation and improve terminal setup handling (6b7b82ea) (@vinhnx)
- Implement Vim mode support with key handling and text operations (e77c61e5) (@vinhnx)
- Implement Vim-style prompt editing engine with core functionality and tests (a722f56b) (@vinhnx)
- Add interactive features for prompt suggestions and job management (89c79954) (@vinhnx)
- Update prompt suggestion docs (aa3aaaae) (@vinhnx)
- Update TODO (d23690fc) (@vinhnx)
- Add OpenAI ChatGPT authentication support and update provider configuration (860a1d65) (@vinhnx)
- Add new GPT-5 Codex models and update related functionality (0cc7a335) (@vinhnx)
- Add client request ID handling and improve responses API integration (fb3220ac) (@vinhnx)
- Scan recent commits for bugs (dbb30d99) (@vinhnx)
- Enhance LLM response processing and metrics logging; update OpenAI model handling (bc8dd6aa) (@vinhnx)
- Enhance OpenAI session handling and update UI status display (caa74f23) (@vinhnx)
- Implement recovery modes for empty responses and adjust tool call handling (e09d51c4) (@vinhnx)
- Add clipboard support for selected text and update default model and theme (439d5cd7) (@vinhnx)
- Refactor tool output handling to support spooled output references and improve clipboard functionality (0339abbd) (@vinhnx)
- Refactor plan block parsing for safe character boundaries and add truncation for tool arguments and stderr previews (4b5a4b12) (@vinhnx)
- Fix read tool config issue (73f5cb6f) (@vinhnx)
- Update TODO (f7f2382f) (@vinhnx)
- Implement dynamic signature generation for task_tracker_create and add tests for payload changes (788215be) (@vinhnx)
- Refactor path and symbol hint extraction in research context collection (539c1ac3) (@vinhnx)
- Update TODO (cc9ee77d) (@vinhnx)
- Fix .gitignore entry for vtcode.toml and reorder configuration fields in vtcode.toml (d5e78f45) (@vinhnx)
- Fix plan mode loading (708b5e99) (@vinhnx)
- Add plan confirmation prompt after draft persistence and enhance error classification (ccd75a8f) (@vinhnx)
- Enhance modal list state management to retain last selection and improve overlay behavior (22f813ce) (@vinhnx)
- Refactor plan mode context handling and update configuration values for consistency (cc191c1d) (@vinhnx)
- Refactor plan validation logic to prefer more complete candidates and enhance related tests (f566129c) (@vinhnx)
- Enhance plan extraction and confirmation flow; add handling for incomplete drafts and improve task tracker management (7a2608f7) (@vinhnx)
- Refactor task tracker signature handling to use duplicate enforcement guard (f93d52a7) (@vinhnx)
- Update TODo (8d522e16) (@vinhnx)
- Update TODO (ac4b4f84) (@vinhnx)
- Refactor plan mode handling and improve session loop logic (86b9b95c) (@vinhnx)
- Update TODO (ffdee65d) (@vinhnx)
- Update circuit breaker configuration and tests for improved resiliency (aa92cd91) (@vinhnx)
- Refactor session layout and rendering logic; consolidate bottom panel handling and improve input height management (e6e93dbf) (@vinhnx)
- Update format (b8d5aee6) (@vinhnx)
- Fix url wrap line (301f62c7) (@vinhnx)
- Refactor session handling and options; streamline overlay requests and keyboard protocol settings (bc30728c) (@vinhnx)
- Refactor input handling; consolidate trigger checks and update input commands (8c46dc45) (@vinhnx)
- Refactor input trigger updates; consolidate trigger checks into a single method (9c7b8562) (@vinhnx)
- Refactor core TUI frame rendering helpers (a59920bf) (@vinhnx)
- Refactor panel layout functions for file palette and history picker; streamline inline area calculations (0a83dff1) (@vinhnx)
- Add note to explore GitHub Copilot OAuth for using Anthropic's Claude (73ed2711) (@vinhnx)
- Add GPT-5.4 Nano and Mini models to OpenAI configuration (982b5c37) (@vinhnx)
- Add OpenAI tool search configuration and implement deferred loading policy (3a56178a) (@vinhnx)
- Add model management documentation and scripts for LLM integration (861bef97) (@vinhnx)
- Add ToolCatalogSource to tool registration and update related files (9db62221) (@vinhnx)
- Update TODO (a2eeed96) (@vinhnx)
- Update TODO.md (a96c1aca) (@vinhnx)
- Add unrelated worktree prompt handling and related functionality (dd59dd65) (@vinhnx)
- Refactor model option handling and validation logic for improved performance and clarity (ed7d837a) (@vinhnx)
- Add OAuth authentication guide and enhance README with OAuth details (0660f149) (@vinhnx)
- Refactor system prompts and skills rendering for improved clarity and efficiency (7ba88e31) (@vinhnx)
- Update README.md (fa33fa70) (@vinhnx)
- Fix formatting in README for installation section (53fccf14) (@vinhnx)
- Update AgentSkills to Skills in README.md (00a37fa0) (@vinhnx)
- Update README.md (3aa6dcc2) (@vinhnx)
- Implement GitHub Copilot authentication and command handling (40da600d) (@vinhnx)
- Add request permission rejection handling and enhance Copilot integration (d713178d) (@vinhnx)
- Fix reasoning formatting in CopilotProvider to ensure proper newline handling (565ae957) (@vinhnx)
- Update notes (a66b4d84) (@vinhnx)
- Implement StdioTransport for JSON-RPC communication and refactor CopilotAcpClient to use it (543dbca1) (@vinhnx)
- Update note (5e7d9492) (@vinhnx)
- Refactor configuration initialization and improve code readability across multiple files; update API key and model settings in configuration. (dd5221bc) (@vinhnx)
- Update code structure for improved readability and maintainability (952b06fe) (@vinhnx)
- Add turn timeout management and recovery logic (bb180e82) (@vinhnx)
- Update TURN timeout recovery to use configured max execution time (e8f27aae) (@vinhnx)
- Refactor tool permission handling and add persistent approval cache (ca53f066) (@vinhnx)
- Implement Ghostty VT backend for terminal emulation and snapshot rendering (2af5a11d) (@vinhnx)
- Refactor code for improved readability and performance; add loading indicators for better user experience (0d20012f) (@vinhnx)
- Add support for custom OpenAI-compatible providers in configuration (8a76debf) (@vinhnx)
- Delete checksums.txt (0d1ffdfd) (@vinhnx)
- Delete skills-lock.json (610fc7b7) (@vinhnx)
- Implement JSON-RPC-over-stdio transport for subprocess agents and refactor authentication methods (090bfc3f) (@vinhnx)
- Enhance ToolCall struct with optional namespace and update related parsing logic (ff97daba) (@vinhnx)
- Refactor RejectConfig documentation and update related tests for clarity (541eac42) (@vinhnx)
- Enhance RejectConfig documentation and update permission rejection logic (d7eb68f0) (@vinhnx)
- Add pod management functionality with model catalog and state handling (c029e115) (@vinhnx)
- Implement normalized streaming for OpenRouter provider and update response handling (b01aeb2f) (@vinhnx)
- Add prompt template support and enhance slash command handling (f23ff5c7) (@vinhnx)
- Prioritize modal key handling over copy-to-clipboard in process_key function (ff0a684a) (@vinhnx)
- Add MiniMax-M2.7 model support and update related configurations (70a0e776) (@vinhnx)
- Add tool invocation lifecycle events and refactor event recording (a67bf917) (@vinhnx)
- Refactor agent session controller and runtime for improved event handling and state management (9adda2cd) (@vinhnx)
- Refactor tool invocation handling and event recording in AgentRunner (9da2d85b) (@vinhnx)
- Refactor tool permission handling and streamline runtime steering integration (f91b0d32) (@vinhnx)
- Add run_standard_stream_attempt function and refactor streaming logic (09a5751f) (@vinhnx)
- Enhance tool permission handling, improve error messaging, and refine path suggestion logic (46521121) (@vinhnx)
- Add file read family tracking and recovery handling in tool outcomes (f88c2a3a) (@vinhnx)
- Implement glob pattern support for file listing and enhance related functionalities (62b7b831) (@vinhnx)
- Implement session mode management and approval caching for shell commands (eda2c70d) (@vinhnx)
- Implement transient UI components and refactor overlay handling (c7622e2d) (@vinhnx)
- Implement prompt cache lineage management and enhance runtime context handling (9e26d8de) (@vinhnx)
- Update session ID and timestamps in current_blocked.md; modify GPU type matching logic and add tests in catalog.rs; change API key and model settings in vtcode.toml (b5a2f974) (@vinhnx)
- Enhance system prompt handling by adding history directives and runtime context checks; add tests for new functionality (c406e88b) (@vinhnx)
- Refactor tool policies and update configuration for Ollama provider (eaaa2b83) (@vinhnx)
- Refactor logging levels from debug to trace for improved verbosity in various components (b8d75afb) (@vinhnx)
- Add history system directives handling and improve session configuration (d7f42900) (@vinhnx)
- Implement centralized logging and tracing system; add flush functionality and guidelines (8903d53c) (@vinhnx)
- Implement interaction support in GeminiProvider with error handling and streaming capabilities (2e1f34d8) (@vinhnx)
- Refactor path handling in tracing and debug logging; simplify flush logic in trace writer (2faf35a7) (@vinhnx)
- Refactor metrics handling to use ToolCatalogCacheMetrics struct for improved clarity and maintainability (495b5476) (@vinhnx)
- Refactor interrupt handling to ensure overlays are closed and improve user experience (6847f30f) (@vinhnx)
- Update TODO (8c9c3d55) (@vinhnx)
- Update system prompt to specify use of plain ASCII icons for markers and bullets (3a8993f0) (@vinhnx)
- Add terminal management features and enhance session handling (8b07cb90) (@vinhnx)
- Update screen_backend.rs (9271147d) (@vinhnx)
- Update config (02384ca5) (@vinhnx)
- Add support for summarized session forks and update related logic (1ddfd7fd) (@vinhnx)
- Add inline prompt suggestion feature to chat composer (b813647f) (@vinhnx)
- Add agent loop contract documentation for VT Code lifecycle and events (f85428b6) (@vinhnx)
- Remove LiteLLM provider and related documentation (ca8889aa) (@vinhnx)
- Update TODO (f7b10332) (@vinhnx)
- Harden agent runner tool error handling (4671b804) (@vinhnx)
- Refactor error handling, optimize token usage, and enhance performance metrics (48af5696) (@vinhnx)
- Add streaming support OpenAI models with normalized event handling (ff76d3af) (@vinhnx)
- Add orchestration for plan-build-evaluate workflow with harness artifacts (6968408b) (@vinhnx)
- Refactor skills system: update documentation, remove version and author fields, streamline validation, and enhance template structure. (e6821250) (@vinhnx)
- Add vtcode-check tool and update related configurations and tests (ae563671) (@vinhnx)
- Remove outdated documentation and improve error messaging format (3f940788) (@vinhnx)
- Remove outdated documentation files and finalize release notes for v0.3.0 (2c2f01d8) (@vinhnx)
- Remove unused dependencies and improve session ID handling in executors (b89fc354) (@vinhnx)
- Refactor recovery handling and improve timeout messaging in execution policy (f0d3f512) (@vinhnx)
- Enhance Responses API support with continuation state management and incremental history handling (0f030292) (@vinhnx)
- Refactor auto mode features and update related configurations (b9843702) (@vinhnx)
- Add diff preview functionality and related utilities (f98e7b7c) (@vinhnx)
- Refactor LLM request retry logic and add previous response chain error handling (0c18e268) (@vinhnx)
- Add support for minimax tool calls and enhance harmony tool parsing (d71bfeb7) (@vinhnx)
- Refactor minimax tool call parsing and add test for invocation without parameters (d50a6bae) (@vinhnx)
- Refactor LLM request retry logic and update provider configuration (b5e47b42) (@vinhnx)
- Refactor tool error handling and retry logic with structured ToolExecutionError (83e1fc97) (@vinhnx)
- Refactor tool error handling to use structured error context (5414b932) (@vinhnx)
- Refactor and consolidate code to reduce duplication across modules (deef3505) (@vinhnx)
- Add subagent controller management to ToolRegistry (c866f9b9) (@vinhnx)
- Add GLM-5.1 model with enhanced capabilities and update references (8b8882f7) (@vinhnx)
- Update .gitignore to include plans directory and tool-policy.json; add Codemod Skill Discovery section to AGENTS.md (67ba167f) (@vinhnx)
- Update TODO (1350cfe5) (@vinhnx)
- Add support for agent references, local agents management, and enhance sidebar rendering (cd51b91d) (@vinhnx)
- Refactor local agents handling and improve UI integration (f3bd979f) (@vinhnx)
- Add parallel batch test for file listing and unified search; refactor list_files to avoid reentrancy (b5b2d2de) (@vinhnx)
- Add minimum turn limit for child agents and update documentation (97eb9a89) (@vinhnx)
- Update theme permissions and synchronize theme selection across components (43361475) (@vinhnx)
- Add theme management and accessibility features (512ad686) (@vinhnx)
- Add documentation for theme-related functions and improve code readability (9c387721) (@vinhnx)
- Enhance terminal restoration on panic and signal handling (d546487f) (@vinhnx)
- Implement scheduled tasks and automation framework (a0c65096) (@vinhnx)
- Add ast-grep integration for code quality checks and rule enforcement (a7b81d0a) (@vinhnx)
- Implement structural search workflows and enhance ast-grep integration (5112cdbb) (@vinhnx)
- Enhance ast-grep documentation and add tests for skill discovery queries (652ab8aa) (@vinhnx)
- Enhance ast-grep documentation and improve pattern syntax guidance (a5a71c9d) (@vinhnx)
- Enhance ast-grep documentation with rule cheat sheets and improve skill discovery tests (0e6a5a6c) (@vinhnx)
- Enhance ast-grep documentation with config cheat sheets and improve skill discovery tests (306f4ab4) (@vinhnx)
- Update TOOL_SPECS.md with strictness details and usage preferences; remove unused stats.rs; enhance SKILL.md and project-workflows.md with ast-grep insights; add tests for skill discovery and structural search flags; refine hints in structural_search.rs. (0cf298f6) (@vinhnx)
- Enhance documentation for custom language setup and tree-sitter integration; add tests for skill discovery by custom language queries. (71b6eadb) (@vinhnx)
- Enhance documentation on language injection and multi-language support; add tests for skill discovery by language queries. (16de0f58) (@vinhnx)
- Enhance ast-grep documentation and language support; add tests for new aliases and inference rules (644d908f) (@vinhnx)
- Enhance documentation for transformation objects and rewriters; add tests for skill discovery by new queries (211e21aa) (@vinhnx)
- Enhance documentation for sgconfig.yml and project workflows; add tests for skill discovery by ruleDirs, testConfigs, libraryPath, and dynamic injected queries. (4c0405d1) (@vinhnx)
- Enhance documentation and refine descriptions for ast-grep commands; normalize exit codes for no matches and findings in structural search. (29867cdc) (@vinhnx)
- Enhance documentation for Rust, TypeScript, TSX, YAML, Ruby, Python, Kotlin, Java, HTML, Go, Cpp, and C catalog examples; provide guidelines for adapting rules to repository-specific conventions. (819298a2) (@vinhnx)
- Enhance documentation for ast-grep project workflows and APIs; clarify project discovery behavior, scan vs run requirements, and performance optimizations (3db5c5e5) (@vinhnx)
- Fix typo in comments for line truncation functions (32ccac09) (@vinhnx)
- Add persistent memory support and enhance instruction appendix functionality (4c73f931) (@vinhnx)
- Refactor signal handling and improve session finalization timeout logic (47828a62) (@vinhnx)
- Refactor credential storage to support encrypted file storage and improve legacy migration handling (d2e97368) (@vinhnx)
- Add model configuration commands and lightweight model support (1458605d) (@vinhnx)
- Refactor transcript link handling and add URL guard feature (5463ab53) (@vinhnx)
- Update TODO.md (cb7ca5c1) (@vinhnx)
- Transform Rust source files (0e4c43d8) (@codemod[bot])
- Implement OAuth support for HTTP MCP providers and enhance CLI commands for login/logout (f6cd0c61) (@vinhnx)
- Add support for Codex app server integration and validation (a2a44dba) (@vinhnx)
- Update config (2d7bb992) (@vinhnx)
- Update TODO (76795760) (@vinhnx)
- Update config (ce0e36d7) (@vinhnx)
- Update TODO (0b58348f) (@vinhnx)
- Add zread badge (6bce27c6) (@vinhnx)
- Refactor tests and update tool policies for improved functionality (2bf10861) (@vinhnx)
- Fix release (ad544b2c) (@vinhnx)
- Fix ast-grep init (0e622cd5) (@vinhnx)
- Fix TUI OAuth cancel without exiting VT Code (ce034434) (@vinhnx)
- Enhance modal rendering with title chrome and divider styles (339be092) (@vinhnx)
- Improve modal list dividers and highlight styling (803a63b5) (@vinhnx)
- Enhance modal rendering with improved styles and background handling (0b4278fb) (@vinhnx)
- Refactor palette rendering to include dividers and improve header styling (a4dde38f) (@vinhnx)
- Add Ghostty VT integration and update dependencies (53431d41) (@vinhnx)
- Add clear background rendering to floating modal and improve tests (c6b1afa0) (@vinhnx)
- Implement reborrow method in SlashCommandContext, update agents command usage, and change OpenAI provider settings in configuration. (2fb54da2) (@vinhnx)
- Update example blocked patterns in vtcode.toml.example (15c69b16) (@vinhnx)
- Enhance /compact command functionality and update configuration settings (8c31b153) (@vinhnx)
- Add freeform default handling to wizard steps and update related prompts (39926d0b) (@vinhnx)
- Fix script (5655308f) (@vinhnx)
- Refactor mock server initialization and error handling in tests; streamline MCP parameter building; update checksum calculation; enhance session rendering in TUI tests; adjust file permissions in configuration. (03d8ccc9) (@vinhnx)
- Update Cargo.toml (1b4763f3) (@Sachin-Bhat)
- Update handlers.rs (bbd13c91) (@Sachin-Bhat)
- Enhance guided /init command skill description, add custom subagent spawning, and update model version in configuration (cfe6a807) (@vinhnx)
- Remove no-console-log test example (a0358d3c) (@vinhnx)
- Add lightweight model palette and update tool policies (393300f1) (@vinhnx)
- Extract terminal detection into a shared workspace crate (73db75f0) (@vinhnx)
- Refactor tool safety shared state handling (53c68714) (@vinhnx)
- Add command auth support for custom providers (9bc0a86e) (@vinhnx)
- Update notes (dd420c71) (@vinhnx)
- Refactor transcript link handling and improve wrapping logic; add fullscreen interaction settings (6c4d8857) (@vinhnx)
- Enhance system prompts and delegation contracts; update sandboxing for macOS preferences; improve skill loading with path and name support; refactor tool handler and JSON schema parsing; add terminal capabilities for editing inputs in TUI. (231c9dc3) (@vinhnx)
- Refactor tool policies by removing unused tools and policies; add Rust trait reasoning guidelines. (a31e4fdd) (@vinhnx)
- Add terminal title setup functionality and related commands (6063928e) (@vinhnx)
- Clarify CGP provider dictionaries (a6b8e2de) (@vinhnx)
- Update TODO (35cc8ccc) (@vinhnx)
- Fix badge duplication in README.md (4ffc53bb) (@vinhnx)
- Migrate to serde-saphyr (c29cdd5d) (@Sachin-Bhat)
- Enhance first-run wizard with startup modes and memory (199d1569) (@vinhnx)
- Fix check script and logging guard regressions (c6781996) (@vinhnx)
- Refactor system prompts and contracts for clarity and conciseness; optimize skill rendering and memory handling; enhance path policy checks; update execution kernel for unified file actions; adjust configuration formatting. (1b1d0c79) (@vinhnx)
- Remove sparse-checkout configuration from release workflow steps (d4be7f53) (@vinhnx)
- Enhance recovery handling and fallback messages; update command sharing options; adjust configuration for model and tool limits. (ebad1810) (@vinhnx)
- Update agent prompt for auto mode rule reviewer and adjust model settings in configuration (7ae71a83) (@vinhnx)
- Add permission request and stop hooks to lifecycle management (5186a3ea) (@vinhnx)
- Refactor permission handling and improve code readability across multiple modules (a28db433) (@vinhnx)
- Implement notification system with /notify command and enhance session start hooks (e1bc7e41) (@vinhnx)
- Add configurable desktop notification backends (bf8c51d8) (@vinhnx)
- Enhance pty command handling, update configuration options, and add new JSON schemas for collaboration and command execution (6cf2dd04) (@vinhnx)
- Refactor vtcode.toml configuration: streamline agent settings, enhance permissions, and update context management (a5e4e440) (@vinhnx)
- Tighten persistent memory extraction and tool policies (f15b43a5) (@vinhnx)
- Add Agent Trajectory Interchange Format (ATIF) support for session trajectory export (0e09ef90) (@vinhnx)
- Update TODO.md (824acefb) (@vinhnx)
- Update notes (e4480f6a) (@vinhnx)
- Fix tool batching and runloop regressions (2593f015) (@vinhnx)
- Consolidate tool fallback and rate-limit parsing (02fdb3c9) (@vinhnx)
- Fix formatting in README.md for clarity (979d991d) (@vinhnx)
- Update note (9f327f92) (@vinhnx)
- Core split (3d12f6b6) (@poelzi)
- Update notes (f3ebfd31) (@vinhnx)
- Refactor slash command messages and update guidance for installation (09028a6a) (@vinhnx)
- Implement status line setup and configuration, deduplicate terminal title items, and update task tracker permission in config (c4b7cb48) (@vinhnx)
- Implement fallback handling for failed tool-free recovery and adjust turn completion logic (4e58ac64) (@vinhnx)
- Implement structured timeout error handling and add tests for timeout scenarios (c214f667) (@vinhnx)
- Update tool policies, enhance exec command preparation, and improve documentation clarity (43c048e4) (@vinhnx)
- Commit (8ed8fb87) (@vinhnx)
- Bump version to 0.50.12 and remove chat-related commands and references (8a8efa6c) (@vinhnx)
- Bump version to 0.50.13 (337fe1ab) (@vinhnx)
- Implement subagent executors and enhance tool intent handling (c550faa2) (@vinhnx)
- Refactor exec session handling and add error reporting and tool search functionality (dc9e150f) (@vinhnx)
- Refactor tool handling and enhance session management for improved output processing (8d8c47ab) (@vinhnx)
- Refactor MCP provider initialization and enhance tool fetching logic (ecff001a) (@vinhnx)
- Refactor compaction thresholds and remove context awareness support from LLM providers (31b771d0) (@vinhnx)
- Update TODO (203ec0f3) (@vinhnx)
- Update tool policies, enhance recovery context previews, and improve structural search handling (a5db2b86) (@vinhnx)
- Enhance tool output handling and summary generation (c10d90fd) (@vinhnx)
- Refactor CLI action resolution and improve header display logic (61f15e54) (@vinhnx)
- Enhance CLI argument parsing and error reporting for workspace validation (77c3efbe) (@vinhnx)
- Add README files for new modules and update existing ones (5c4384be) (@vinhnx)
- Add subagent constants, discovery, model resolution, prompt handling, and types (63230b30) (@vinhnx)
- Add tests for transcript rendering, Vim mode functionality, and wizard interactions (1505cbfa) (@vinhnx)
- Update code structure for improved readability and maintainability (39917c56) (@vinhnx)
- Add persistent memory tests for memory operations and classification (a0e0a343) (@vinhnx)
- Refactor code structure for improved readability and maintainability (3ffdf8a4) (@vinhnx)
- Refactor code for improved readability and consistency; update default model in configuration (d42726a2) (@vinhnx)
- Update theme in configuration to 'ciapre-blue' (c12bdf36) (@vinhnx)
- Add tool outcomes handling for response content and subagent memory (1327986f) (@vinhnx)
- Add session archive tests for persistence and message handling (311bd727) (@vinhnx)
- Refactor tool names to use constants from the tools module for consistency (44181fa5) (@vinhnx)
- Add runtime tests for Copilot functionality in llm_request module (d038b0cf) (@vinhnx)
- Add comprehensive tests for tool execution and context handling (0d3b4776) (@vinhnx)
- Refactor code for improved readability and consistency in copilot_runtime, components, and skills modules (b2cc57e0) (@vinhnx)
- Add tests for structural search functionality and request validation (f4649c89) (@vinhnx)
- Add tests for tool execution and sandbox runtime behavior (cbf7effb) (@vinhnx)
- Refactor UI command item creation for improved readability and maintainability (0db3b4d4) (@vinhnx)
- Update tool policies, fix keyboard shortcuts, and enhance documentation (8b91a421) (@vinhnx)
- Add websocket active response error handling and tests (a784ca34) (@vinhnx)
- Refactor tool policies and enhance lifecycle hook payloads to include tool call ID (f27a37cd) (@vinhnx)
- Add support for new session trigger and refactor session initialization logic (7fe243b0) (@vinhnx)
- Add grouping functions for MCP tools by provider and update related logic (b58a72f9) (@vinhnx)
- Add warning handling for session archive persistence and finalization (7341d25e) (@vinhnx)
- Update descriptions for agent parameters and built-in tool registrations (69e86421) (@vinhnx)
- Update command preview notice to include sharing instructions (4588b09c) (@vinhnx)
- Refactor diagnostics and UI interaction to enhance instruction context handling (f1bcb2d4) (@vinhnx)
- Refactor JSON schema handling and enhance tool specifications (c719871f) (@vinhnx)
- Add meta field to MCP elicitation request and response structures (914d59b8) (@vinhnx)
- Add notification condition for desktop notifications delivery (559bba94) (@vinhnx)
- Add default target for docs.rs in Cargo.toml files (12a5d068) (@vinhnx)
- Refactor code for improved readability and consistency in multiple files (7cc15514) (@vinhnx)
- Remove outdated TODO items from project documentation (58aec7ee) (@vinhnx)
- Update deps (bd9e90ef) (@vinhnx)
- Update TODO (92336dc0) (@vinhnx)
- Update tool execution status display and refine UI handling (f77c7053) (@vinhnx)
- Implement pending user prompt handling and enhance status rendering in TUI (2c4dae7b) (@vinhnx)
- Advertise ACP commands (86c00c9c) (@vinhnx)
- Update Anthropic provider to support Claude Opus 4.7 with new task budget and reasoning effort features (c0824531) (@vinhnx)
- Add budget limit continuation support and related metadata handling (b46d4e3c) (@vinhnx)
- Refactor Anthropic provider to support adaptive thinking and update reasoning effort handling (01b3b1b5) (@vinhnx)
- Refactor OpenCode providers to support model ID resolution and enhance compatibility (0a5bf2c1) (@vinhnx)
- DeepSeek V4 Models identifiers and descriptions to reflect version changes (be3638ba) (@vinhnx)
- Update tests (896b1494) (@vinhnx)
- Update (a4c12b76) (@vinhnx)
- Update (e3590bd5) (@vinhnx)
- Add DeepWiki badge to README (7b21006a) (@vinhnx)
- Update config (59a5f22c) (@vinhnx)
- Add support for spawning background subprocesses and related configurations (17665666) (@vinhnx)
- Refactor configuration loading and improve error handling across modules (b7be8464) (@vinhnx)
- Add deepseek-v4-pro:cloud under Ollama provider (512f260b) (@vinhnx)
- Fix ast-grep dependencies install (21585180) (@vinhnx)
- Enhance OpenAI Response's API websocket provider logic and tests; clarify documentation and comments. (7368a946) (@vinhnx)
- Remove keyring dependency and add platform-specific keyring stores (fef8f0a2) (@vinhnx)
- Refactor tool execution and configuration; add tool call delay tests and enhance agent capabilities (a8ce67f7) (@vinhnx)
- Update AGENTs guide (c5b71a50) (@vinhnx)
- Update document about lacking support of Windows build (1757d58d) (@vinhnx)
- Implement slash command handling and associated models; add tests for command parsing and outcomes (2d07c12f) (@vinhnx)
- Add status line and terminal title setup commands (edca2bfb) (@vinhnx)
- Add support for inline loop actions and enhance diagnostics commands (34fc9261) (@vinhnx)
- Decompose TUI runloop and slash-command modules into smaller units (02d94320) (@vinhnx)
- Add interview context and payload handling for Plan Mode (d0516994) (@vinhnx)
- Add recovery preview and response handling for assistant interactions (d0b6e8e8) (@vinhnx)
- Refactor canonical_tool_name to return &str and update related usages (0dbef927) (@vinhnx)
- Refactor agent_type handling to use String instead of Cow and optimize message content processing (3901951c) (@vinhnx)
- Refactor code for readability and maintainability; streamline tool loop limit checks and enhance plan response handling (3d222d8e) (@vinhnx)
- Enhance Clippy lints for better safety and correctness checks (8b03a3e1) (@vinhnx)
- Replace #[allow(dead_code)] with #[expect(dead_code)] in multiple files to improve dead code handling (b01fb753) (@vinhnx)
- Add new tools to tool policy and adjust policies; update Clippy lints and test expectations (f9c6c5df) (@vinhnx)
- Update lints and improve code quality across multiple modules (ca987323) (@vinhnx)
- Add Gemini 3.5 Flash model (0509cebf) (@vinhnx)
- Remove deprecated OpenRouter models and update data structures for better type handling (06ccb60f) (@vinhnx)
- Refactor AgentRunner initialization and remove unused TUI session files (7da13a76) (@vinhnx)
- Remove integration example and MCP chat adapter files; update backend integration with ChatMessage interface (31e6dafc) (@vinhnx)
- Refactor tool outcome handlers and improve recovery guidance (1af72c93) (@vinhnx)
- Refactor token usage tracking and simplify recovery logic in turn loop (a284faf9) (@vinhnx)
- Update brew (36328c56) (@vinhnx)
- Update config (632d5235) (@vinhnx)
- Refactor completion handling and enhance tool recovery logic (ce916001) (@vinhnx)
- Optimize string allocations and improve performance in various modules (2300d49f) (@vinhnx)
- Refactor environment variable handling with env_lock; remove unsafe code in tests (407414e0) (@vinhnx)
- Update tool policies to allow additional actions and modify reasoning effort handling (2266862a) (@vinhnx)
- Update config (e95a9d64) (@vinhnx)
- Update TODO (f679482a) (@vinhnx)
- Update vtcode (9aa12ed8) (@vinhnx)
- Refactor enums to derive Copy trait for improved performance and memory efficiency (bf7f239b) (@vinhnx)
- Optimize performance and memory usage across various modules by utilizing `Vec::with_capacity`, `entry` API, and reducing unnecessary clones. Mark cold paths with `#[cold]` for better inlining control. Refactor string manipulations for efficiency and clarity. (3950f5f7) (@vinhnx)
- Refactor formatting functions to use `vtcode_commons::formatting::collapse_whitespace` for whitespace handling and optimize string concatenation using `write!` for improved performance. (0273970c) (@vinhnx)
- Fix release script (6636c6bc) (@vinhnx)
- Add support for Qwen models in the model ID system (7445df05) (@vinhnx)
- Update Anthropic model to Claude Opus 4.8 and adjust related configurations (6ec1aecf) (@vinhnx)
- Add compaction support to response structures and documentation (4ad4f344) (@vinhnx)
- Update TODO (c4e2d4a0) (@vinhnx)
- Update changelog (9ba9979b) (@vinhnx)
- Add support section with donation link in README (cfd63e1d) (@vinhnx)
- Refactor LLM provider handling and improve error context management (b2cd624a) (@vinhnx)
- Remove deprecated DeepSeek V3.2 model references and update related tests (5ce314cf) (@vinhnx)
- Refactor model IDs and descriptions; remove deprecated models and update defaults (9b967eeb) (@vinhnx)
- Add LlamaCpp provider support and related configurations (12c5982c) (@vinhnx)
- Add support for managing local inference servers and new model integration (82ae1a5c) (@vinhnx)
- Add Ctrl+M shortcut to open model picker and update related tests (73538876) (@vinhnx)
- Add donation command and update related UI prompts (fd1a354d) (@vinhnx)
- Fix minimum Ollama version for Responses API and update model fields to be optional (4dc547e3) (@vinhnx)
- Add PR code review agent and related instructions (4f443419) (@vinhnx)
- Add PR code review guidelines to AGENTS.md (14d0e5cc) (@vinhnx)
- Add workspace structure and build guidelines to AGENTS.md (5009acb7) (@vinhnx)
- Refactor AGENTS.md to streamline rules and workspace structure (0ad20837) (@vinhnx)
- Add AGENTS.md files for new modules and update .gitignore to include memory directory (2be6f557) (@vinhnx)
- Update LM Studio provider to support new models and enhance API integration (86671657) (@vinhnx)
- Refactor file operations to use async Tokio methods for improved performance (e4582be4) (@vinhnx)
- Add criterion for benchmarking and improve text editing functions (dc22c528) (@vinhnx)
- Fix dependency (208d6396) (@vinhnx)
- Update TODO (2cb0c40b) (@vinhnx)
- Update dependencies and enhance input management with undo/redo functionality (1b932528) (@vinhnx)
- Refactor LLM provider implementations to remove backend_kind method and improve code readability (fc12014f) (@vinhnx)
- Fix installer filename formatting (86e9fcc7) (@uiYzzi)
- Update README (f2b4202b) (@vinhnx)
- Update TODO (017be561) (@vinhnx)
- Add help modal using ratatui-cheese for keyboard shortcuts (87f3630e) (@vinhnx)
- Update UI highlight symbols and styles across multiple components; change theme to "vitesse-black". (41503473) (@vinhnx)
- Refactor input handling and styles across various components; unify search field rendering; update theme to ciapre-blue. (33a910d6) (@vinhnx)
- Audit and refactor (9a26456c) (@vinhnx)
- Update API key and image sizes in README.md (b7c60aa6) (@vinhnx)
- Simplify quick start instructions for VT Code (3dff646f) (@vinhnx)
- Add vtcode update command to README (974cfd25) (@vinhnx)
- Enhance images in README with rounded corners; update TODO and configuration settings (87a81d8d) (@vinhnx)
- Refactor tool name handling to use CompactStr for improved memory efficiency (8c1c7a30) (@vinhnx)
- Add openrouter code generation and macros for string newtypes (b875258a) (@vinhnx)
- Implement core terminal functionality with screen management, input processing, and device reporting (ae290963) (@vinhnx)
- Refactor file operations and enhance error handling (e09bf2b5) (@vinhnx)
- Add unified style bridging between anstyle and ratatui (47e14082) (@vinhnx)
- Remove Ghostty VT runtime integration and related tests; adjust terminal size in tests. (2218ca09) (@vinhnx)
- Update TODO (8f62b657) (@vinhnx)
- Feat/top level agents (533af16b) (@kernitus)
- Refactor file exclusion logic and enhance directory walking utilities (21119580) (@vinhnx)
- Enhance documentation and configuration for GhosttyCore integration; update related tests and clean up lint settings (0d210b1e) (@vinhnx)
- Update tree-sitter grammar handling and enhance unified search parameters for rewrite workflows (4eaa8c22) (@vinhnx)
- Add linting rules for Kotlin, Python, Ruby, and Rust to enforce best practices (f44d5691) (@vinhnx)
- Add linting rules for Python, Rust, TypeScript, and TSX to enforce best practices (e049bc14) (@vinhnx)
- Refactor and enhance linting rules for Python and Kotlin; add new rules for identity checks and update metadata for existing rules (acb69e45) (@vinhnx)
- Add rule snapshots for various code quality improvements (4ba0b8dd) (@vinhnx)
- Add tests for relational rule YAML generation and validation (85c1f44f) (@vinhnx)
- Add support for C and C++ in local parser checks; update tests accordingly (e51eceee) (@vinhnx)
- Refactor StructuralSearchRequest to use serde_json types directly; remove unused validation functions (cbb3d734) (@vinhnx)
- Add new rules and updates for various languages; enhance JSON schema for unified search parameters (db75ccf4) (@vinhnx)
- Remove unused dependencies, enhance documentation, and update various rules for improved code quality and consistency across languages. (3cd68c2a) (@vinhnx)
- Remove stale lint expectations that are now fulfilled (9aa04c7d) (@vinhnx)
- Update TODOs (c47dd4fb) (@vinhnx)
- Update TODO (d41ae5ed) (@vinhnx)
- Refactor tool registration to use distributed slices and update README contributors script (1f1c6978) (@vinhnx)
- Remove outdated documentation references and clean up project files (ebdfe55f) (@vinhnx)
- Update ignore (114a62ec) (@vinhnx)
- Update TODO (b7d216f1) (@vinhnx)
- Fable (8781f2a4) (@vinhnx)
- Update README (7565cbac) (@vinhnx)
- Update README (7f4186dd) (@vinhnx)
- Ignore vtcode plan artifact (20736d87) (@vinhnx)
- Feat/merge modes & agents (3e743390) (@kernitus)
- Fix/671 chatgpt OpenAI requests (071c9e64) (@kernitus)
- Update TODO (d54a296f) (@vinhnx)
- Update README (27a96ce8) (@vinhnx)
- Add release notes command and display on startup after updates (8f6848fd) (@vinhnx)
- Refactor model provider mappings and remove deprecated models (9ead7f9c) (@vinhnx)
- Update TODO (c272494b) (@vinhnx)
- Fix updater (faf13b81) (@vinhnx)
- Update TODO (b0a2c9f3) (@vinhnx)
- Add MiMo authentication method support and related configurations (b23ac5d1) (@vinhnx)
- Update TODO (ddf079ae) (@vinhnx)
- Add MiMo authentication method support and update UI constants (f80f806d) (@vinhnx)
- Remove terminal implementation files and related functionality (419d0dea) (@vinhnx)
- Refactor rules and snapshots for various languages; update no-malloc-free, no-unnecessary-let, and improve error handling in Go; enhance snapshot tests for better coverage. (5e4b1160)
- Consolidate collaboration tool specs into utility tool specs and remove redundant files (e6c80dac)
- Refactor tool policies and retry mechanisms; update error handling and improve middleware structure (462b4e85)
- Add pattern detection, workflow optimization, and middleware for tool execution (be7aec56)
- Remove vtcode-tools crate and consolidate functionality into vtcode-core (23ffa1a6)
- Remove vtcode-lmstudio crate and consolidate related functionality into vtcode-core (72ab170d)
- Refactor error handling and savings estimation in memory and summarization modules (fc613ce4)
- Enhance error handling and improve regex pattern validation across multiple modules (1cfc9220)
- Refactor code to improve readability and maintainability (7cdfe0cb)
- Refactor tool policy and documentation; remove dead code and unused features; merge related crates for efficiency; clean up test duplicates and re-export paths. (3c94d63e)
- Add vtcode-tool-types crate with shared runtime types for tool system (95e49b8b)
- Add CLI tool templates and instructions for VT Code integration (da316c63)
- Implement Open Responses specification with request and response models (6bca4c1f)
- Remove vtcode-vim module and add Vim state management in vtcode-ui (28688001)
- Update documentation and refactor crates for clarity and consistency (053b4e59)
- Refactor layout handling and improve visual dirty marking in session management (998dda9c)
- Refactor terminal capability checks for rich Unicode support and update default theme to mono (e4c0c663)
- Refactor tool response handling to include read-once guidance and improve file summarization hints (b18773aa)
- Add read-after-write guard and enhance file tracking in tool execution (fff9a2f9)
- Add guideline to avoid using emoji in responses (3847a5cb)
- Refactor contract lines and parameter schemas for consistency and clarity (b57b516d)
#### Performance
- Remove redundant clones and allocations (c21dfa9d) (@vinhnx)
- Use write! macro instead of format! with push_str (b629f8a9) (@vinhnx)
- Add Vec with_capacity for known-size allocations (8136ed79) (@vinhnx)
- Eliminate redundant clones in config loader (5f3ae061) (@vinhnx)
- Use .to_string() directly for Display types instead of .as_str().to_string() (92286096) (@vinhnx)
- Remove unnecessary clone() on Copy types (adb894cb) (@vinhnx)
- Optimize string formatting and use unwrap_or_default (0607a8da) (@vinhnx)
- Use write!/writeln! macros in diff_renderer (325132f5) (@vinhnx)
- Use write!/writeln! macros in tools and prompts modules (a99c47b2) (@vinhnx)
- Use .to_string() instead of format! for context_size (460b11ae) (@vinhnx)
- Use write! macro in core token management modules (8cf17549) (@vinhnx)
- Use write! macro in exec modules and utils (ca576bc6) (@vinhnx)
- Use write! macro in llm/token_metrics (9e7cc8ac) (@vinhnx)
- Use write! macro in metrics module (e1cd82b9) (@vinhnx)
- Optimize memory allocations and pre-allocate buffers in various modules (e12629b8) (@vinhnx)
- Optimize rate limiting with a read-lock fast path and refactor tool execution retry delays using constant values. (a8af7997) (@vinhnx)
- Cache session header lines and queued input previews to optimize TUI rendering performance and remove outdated content from TODO.md. (cf87bc80) (@vinhnx)
#### Refactors
- Update configuration for provider and model changes; enhance spinner display (c936153e) (@vinhnx)
- Modularize agent bootstrap components (74749e15) (@vinhnx)
- Update links in TODO.md and clean up rendering logic in turn.rs (8c58fd3c) (@vinhnx)
- Centralize openai tool serialization (90a1ebaf) (@vinhnx)
- Update provider settings and adjust demo dimensions (cde1e874) (@vinhnx)
- Remove indicatif dependency and related spinner code (d78e10e4) (@vinhnx)
- Remove slash command suggestion functionality (98e70000) (@vinhnx)
- Replace HashMap with BTreeMap in MCP configuration (e8e7801f) (@vinhnx)
- Enhance slash command suggestion management (f1050f0f) (@vinhnx)
- Remove slash command suggestion functionality (aa775644) (@vinhnx)
- Replace HashMap with BTreeMap in MCP configuration (eeda4c15) (@vinhnx)
- Reorder available tools and update policies (cfdd507e) (@vinhnx)
- Update TODO with project cleanup tasks and change script permissions (48f71cea) (@vinhnx)
- Update TODO with project cleanup tasks and change script permissions (7121e88f) (@vinhnx)
- Format dependencies for better readability (8a3f6738) (@vinhnx)
- Simplify reasoning effort handling in AgentRunner (49d94ec2) (@vinhnx)
- Add default_system_prompt function for system prompt retrieval (e03993b5) (@vinhnx)
- Replace DEFAULT_SYSTEM_PROMPT usage with default_system_prompt function (ebde002a) (@vinhnx)
- Prune unused dependencies (7bd6b964) (@vinhnx)
- Migrate acp-client to vtcode-acp-client and update dependencies (f76eff48) (@vinhnx)
- Render tool outputs with ratatui panels (c96008c5) (@vinhnx)
- Remove outdated system prompts for Coder, Explorer, Orchestrator, and Prompt Refiner (8f08f82a) (@vinhnx)
- Remove unused tools and simplify tool policies (446bd29b) (@vinhnx)
- Use unsafe blocks for environment variable manipulation in tests (e42ea782) (@vinhnx)
- Rename extension from "VTCode Companion" to "VTCode" and update CHANGELOG (2c5f17bd) (@vinhnx)
- Move display_interrupt_notice function to improve code organization (c67bfd69) (@vinhnx)
- Update configuration for LLM provider and model settings (8c297c56) (@vinhnx)
- Update LLM provider and model configurations (ac6e3aa7) (@vinhnx)
- Improve tool summary rendering and clean up unused code (6a2d9bd0) (@vinhnx)
- Rename RUN_TERMINAL_CMD to RUN_COMMAND for consistency (946fa1cf) (@vinhnx)
- Update LLM provider and model configurations (7a180237) (@vinhnx)
- Clean up whitespace and improve code readability (f887004c) (@vinhnx)
- Enhance conversation compression logic and message truncation (6483ef86) (@vinhnx)
- Update tool policies to allow more actions and improve asset URL resolution (ffc0c279) (@vinhnx)
- Update tool policies for curl and apply_patch, and improve error messages in update checker (0ee030e2) (@vinhnx)
- Improve code formatting and readability in various files (2bd2efeb) (@vinhnx)
- Modernize TUI rendering and improve diff visualization (9ab0c1d5) (@vinhnx)
- Improve sandbox configuration and event logging performance (b4bd8cf4) (@vinhnx)
- Optimize file metadata retrieval and reduce unnecessary system calls (0f9fdc0c) (@vinhnx)
- Remove unused IdeContextBridge and clean up session initialization (87eef0a4) (@vinhnx)
- Update docs.rs URL and improve response handling in release script (50deb6e8) (@vinhnx)
- Remove npm package support and update installation instructions (071feb34) (@vinhnx)
- Improve logging for AST-grep engine initialization (857281d8) (@vinhnx)
- Remove unused imports and streamline timeout error handling (94805898) (@vinhnx)
- Update extension ID and name for consistency (38d5f3b3) (@vinhnx)
- Update tool policy and improve command handling; streamline error messages and enhance telemetry logging (81b0ccc7) (@vinhnx)
- Update tool policies, exclude zed-extension from workspace, and upgrade zed_extension_api dependency (ddf43255) (@vinhnx)
- Remove unused zed-extension files and grammars (542832ce) (@vinhnx)
- Replace assert_fs::prelude with tempfile::tempdir in tests (2d997fdf) (@vinhnx)
- Clean up test module by removing unused imports and structures (d44e1d9d) (@vinhnx)
- Remove wasm32-wasip2 component from toolchain configuration and clean up test imports (68ab5583) (@vinhnx)
- Remove unused app constant and update elicitation capability handling (829d28f0) (@vinhnx)
- Update tool policies and improve MCP tool handling (94abbb67) (@vinhnx)
- Streamline MCP tool handling and validation logic (af974945) (@vinhnx)
- Restructure zed-extension to modular architecture with comprehensive error handling and caching (a7e86de3) (@vinhnx)
- Integrate CommandBuilder throughout commands module (f07a82bd) (@vinhnx)
- Remove AST-grep tool and related components (aefcd24d) (@vinhnx)
- Implement styling refactor from guide - centralize color/style management (d271d6a3) (@vinhnx)
- Implement styling suggestions from STYLING_REFACTOR_GUIDE (94fdbf3b) (@vinhnx)
- Improve styling consistency with bold_color() and ColorPalette (9155223f) (@vinhnx)
- Implement central style helpers and diff color palette (d2cdfe2f) (@vinhnx)
- Extract input history navigation logic fix in InputManager (a65242f5) (@vinhnx)
- Phase 2 Step 1 - add manager fields to Session struct (6bdb3649) (@vinhnx)
- Phase 2 Step 2 - add manager sync helper methods (6a304970) (@vinhnx)
- Phase 2 Step 3 - migrate clear_input() and reset_history_navigation() (e8a6b327) (@vinhnx)
- Phase 2 Step 4 - migrate remaining input methods and word navigation (6b34988e) (@vinhnx)
- Remove references to ast_grep_search from documentation and tool policies (568765af) (@vinhnx)
- Remove unused audit log and history navigation methods (29c7b31e) (@vinhnx)
- Update command execution in tests and remove unused imports (d7c37d40) (@vinhnx)
- Update InlineTextStyle to include bg_color and effects in snapshot tests (59645d8f) (@vinhnx)
- Replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization (cf7eae04) (@vinhnx)
- Remove obsolete documentation files for `vtcode-bash-runner`, `vtcode-config`, `vtcode-core`, `vtcode-exec-events`, and `vscode-extension` as part of the extraction strategy to streamline the codebase and focus on standalone crate development. (2d3761b5) (@vinhnx)
- Extract tool pipeline into and add — reduce run loop complexity (b4c5ceb4) (@vinhnx)
- Organize documentation into docs/phases and docs/scroll subdirectories; consolidate PHASE5 and SCROLL artifacts for better maintainability (7361dd72) (@vinhnx)
- Replace dissimilar with optimized Myers diff algorithm (899884d3) (@vinhnx)
- Streamline loop detection logic and improve non-interactive handling (c1870eda) (@vinhnx)
- Streamline command execution error suggestions and implement unified run command executor (ac306a47) (@vinhnx)
- Rename bash to shell and mark run_pty_cmd as deprecated (4a60d3ab) (@vinhnx)
- Simplify execute_shell_command to skip conversion layer (92587224) (@vinhnx)
- Dramatically simplify execute_shell_command (e5ba455b) (@vinhnx)
- Simplify execute_shell_command further (f6ae4643) (@vinhnx)
- Use pattern matching in execute_shell_command for clarity (50945608) (@vinhnx)
- Eliminate wrapper layer in execute_shell_command (a829160c) (@vinhnx)
- Remove sandbox functionality and streamline shell command (b6844d6c) (@vinhnx)
- Remove static default editor, rely on try_common_editors for fallback (9d7e2175) (@vinhnx)
- Rename `run_terminal_cmd` to `run_pty_cmd` across documentation, examples, and tests. (2bc4a71e) (@vinhnx)
- Clean up unused imports (32599105) (@vinhnx)
- Rename `legacy` module to `file_helpers` and fix critical `edit_file` bugs related to newline handling, matching, and trailing newlines. (ab305f8e) (@vinhnx)
- Remove redundant reasoning handling, clarify intent (d33d5ba6) (@vinhnx)
- Update ThinkingSpinner struct visibility for better encapsulation (b3bae451) (@vinhnx)
- Streamline code formatting and improve readability across multiple files (b6e29def) (@vinhnx)
- Optimize diff rendering and suppression logic (0d8cb5b0) (@vinhnx)
- Extract common provider helpers to reduce duplication (a1371ade) (@vinhnx)
- Add parse_tool_call and map_finish_reason helpers (2553defd) (@vinhnx)
- Add validate_request_common helper (e3388c98) (@vinhnx)
- Extract serialize_messages_openai_format helper (2a0d282f) (@vinhnx)
- Update message type handling and improve system prompt instructions (c21d1345) (@vinhnx)
- Optimize core modules for code quality and allocations (442f0798) (@vinhnx)
- Optimize linting and code completion modules (0e10b64b) (@vinhnx)
- Optimize completion learning modules with .into() patterns (64bb3040) (@vinhnx)
- Remove unused tools from tool policies (95134f6b) (@vinhnx)
- Improve code readability by simplifying conditional statements and updating deprecated usages across multiple modules (9f6b2d6b) (@vinhnx)
- Enhance reasoning model support and update tool policies; remove deprecated Moonshot models (069576ff) (@vinhnx)
- Update model configurations to use OpenRouter for Moonshot models; remove deprecated entries and enhance model support (aa6440f8) (@vinhnx)
- Introduce warning and error handling methods in AgentRunner; streamline tool failure logging and path normalization utilities (4e1a1887) (@vinhnx)
- Optimize core agent execution and consolidate utility modules (151963bd) (@vinhnx)
- Implement API failure tracking with exponential backoff; optimize tool caching and navigation (4fad6ceb) (@vinhnx)
- Improve code formatting and structure across multiple files (036b1a5c) (@vinhnx)
- Update tool policy and enhance tool validation (7c8fada8) (@vinhnx)
- Unify token budget constants and improve context management (7d25f378) (@vinhnx)
- Streamline conditional checks and improve code readability (44aeb8ec) (@vinhnx)
- Improve code clarity and consistency in multiple modules (b80071f9) (@vinhnx)
- Enhance McpAllowListConfig structure and improve rule definitions (7432220a) (@vinhnx)
- Implement tool denial handling in agent runner (aefcd39f) (@vinhnx)
- Enhance tool policy and prompt clarity (b7cabbda) (@vinhnx)
- Implement tool execution retry logic and enhance error handling (452d52f1) (@vinhnx)
- Enhance system prompts with improved tool safety and execution guidelines (8b43c83d) (@vinhnx)
- Enhance context management and prompt generation (4083e210) (@vinhnx)
- Enhance PTY command output summarization and follow-up handling (20fac650) (@vinhnx)
- Update tool policy and streamline follow-up handling (60759a71) (@vinhnx)
- Remove deprecated tools and update tool policies (215de7ed) (@vinhnx)
- Update tool policy and improve middleware handling (de61977e) (@vinhnx)
- Simplify ToolCallUpdateFields initialization (55e9ffe1) (@vinhnx)
- Streamline initialization and error handling in various modules (a5d054e5) (@vinhnx)
- Simplify line style selection logic in tool output (23c7e388) (@vinhnx)
- Enhance loop detection logic and add tests (0449f546) (@vinhnx)
- Remove outdated vibe_tooling_mapping documentation (c548f341) (@vinhnx)
- Remove logging statements from UI interaction and session handling (49ac3da3) (@vinhnx)
- Streamline error handling and conditional checks in various modules (a884c396) (@vinhnx)
- Update GPT-5.2 model identifiers and documentation (da05f18d) (@vinhnx)
- Update configuration and documentation for improved clarity and performance (c3cd405c) (@vinhnx)
- Enhance context trimming and session management features (541743af) (@vinhnx)
- Enhance Z.AI provider error handling and API key validation (2c640a0d) (@vinhnx)
- Standardize LLMError structure across providers (1fe1038a) (@vinhnx)
- Update LLMError handling in turn_processing (04ff4c98) (@vinhnx)
- Remove unnecessary whitespace in diff and test files (a6a0ae22) (@vinhnx)
- Remove router configuration and related core logic (2b939c2e) (@vinhnx)
- Streamline widget block creation and layout definitions in TUI components (e903d45a) (@vinhnx)
- Migrate testing from `cargo nextest` to `cargo test` and enhance Anthropic LLM configuration with new parameters. (62a8e103) (@vinhnx)
- Clean up whitespace and formatting across multiple files for improved readability (77f3a51a) (@vinhnx)
- Apply clippy fixes for code quality improvements (07dec14b) (@vinhnx)
- Simplify configuration handling and update tool permissions in multiple files (8fe94ae8) (@vinhnx)
- Simplify error handling in dotenv loading (fa46b86c) (@vinhnx)
- Streamline toolset by merging agent diagnostics and removing deprecated tools (b3a820ab) (@vinhnx)
- Update reasoning labels for clarity in justification and session headers (b10c546d) (@vinhnx)
- Restructure agent turn execution with new guard, context, and tool outcome modules, removing old loop detection, and updating LSP tools. (ce2d082a) (@vinhnx)
- Consistently use `adaptive_trim` with `pruning_ledger` across all proactive token budget guards. (6eb56b13) (@vinhnx)
- Introduce a dedicated interaction loop for centralized user input and turn flow, updating session and context management. (7263960a) (@vinhnx)
- Improve error message for missing MCP tools with installation instructions (3153add2) (@vinhnx)
- Remove token budget management and related token estimation/truncation components, and add associated documentation and verification scripts. (37c65171) (@vinhnx)
- Overhaul TUI, tool policy, and context management, adding new documentation and tests. (5999e45f) (@vinhnx)
- Refine tool policies by removing unused tools, changing several to prompt-based, and making `wrap_text` test-only. (a0e12516) (@vinhnx)
- Remove unused agent diagnostic tools from TODO documentation (ace1d65f) (@vinhnx)
- Update tool policies to prompt-based for MCP time functions and improve session handling with cancellation support (8b8cfd8a) (@vinhnx)
- Enhance analysis command to support multiple analysis types and improve error handling (5dcb5deb) (@vinhnx)
- Clean up unused imports and improve webhook handling (b2e0d08b) (@vinhnx)
- Improve Linux checksum handling and release verification in scripts (99686284) (@vinhnx)
- Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management (47be9b07) (@vinhnx)
- Extract progress update guard and elapsed time updater to `progress.rs` and add `PlaceholderSpinner::force_refresh`. (12f1c890) (@vinhnx)
- Update symbol name extraction to use `ChildByField` and add Rust language test. (f46b334b) (@vinhnx)
- Simplify nested conditional logic with chained `&& let` patterns (61cb1e0a) (@vinhnx)
- Clean up code formatting and improve readability across multiple files (4c233b4e) (@vinhnx)
- Update tool policies, enhance subagent cleanup, and improve documentation (dfb0eb7c) (@vinhnx)
- Update tool policies to allow write_file and unified_file actions; remove redundant error logging (e612fcc6) (@vinhnx)
- Clean up code formatting and improve readability in multiple files (ecc34001) (@vinhnx)
- Streamline code by simplifying conditional checks and improving output handling (1047a0bb) (@vinhnx)
- Remove unused LLM provider implementations (85df715a) (@vinhnx)
- Update reasoning color and style for improved readability and placeholder effect (232fe666) (@vinhnx)
- Inline table row rendering logic, remove duplicate parameters, and clean up the TODO list. (16e46b4f) (@vinhnx)
- Remove unnecessary cloning and assignment of `_updated_snapshot`. (001cee41) (@vinhnx)
- Replace map_or with is_none_or for improved clarity in MCP tool filtering (46214568) (@vinhnx)
- Simplify tracing initialization with unwrap_or_default (198a3b5d) (@vinhnx)
- Simplify conditional checks for context awareness in prompt building (78e83e8a) (@vinhnx)
- Update path parameter types from PathBuf to Path for consistency (acc0f07c) (@vinhnx)
- Update TODO list with improved queue messages UI and handling (04980f8b) (@vinhnx)
- Update test assertions for clarity and accuracy (381f2874) (@vinhnx)
- Enhance history navigation and update inline event handling (61f7ab27) (@vinhnx)
- Improve history navigation and update input handling (cfa656d2) (@vinhnx)
- Implement Chain-of-Thought monitoring and context anxiety management patterns (79dfbc20) (@vinhnx)
- Update thinking budget constants and enhance extended thinking configuration (03566c07) (@vinhnx)
- Streamline syntax highlighting by introducing a dedicated module and optimizing theme management (900e4ceb) (@vinhnx)
- Optimize inline event handling and improve command safety checks (ce6eb7a4) (@vinhnx)
- Enhance log event filtering and improve user message styling in TUI (7ad87591) (@vinhnx)
- Optimize string handling with dereferencing in multiple files (7c4fbc54) (@vinhnx)
- Use AsRef trait for string conversion in command rendering (3c63a527) (@vinhnx)
- Adjust output thresholds and preview line counts for improved token efficiency (85fce4ca) (@vinhnx)
- Remove tui-syntax-highlight dependency and streamline syntax highlighting implementation (4ba79b07) (@vinhnx)
- Simplify codebase by auditing markdown.rs, removing unused tests, and standardizing effort parameter in tool calls (8506368c) (@vinhnx)
- Rename default method to default_cache and update default implementations for various structs (9828b3c9) (@vinhnx)
- Integrate clean_reasoning_text function to streamline reasoning text handling across multiple modules (0e6ca527) (@vinhnx)
- Increase max_conversation_turns to 150 and streamline allowed_tools format; add auto_exit_plan_mode_attempted to context for improved plan mode handling (b58ee477) (@vinhnx)
- Update exit_plan_mode policy to prompt; enhance OpenResponsesProvider with version handling and reasoning content support (65d2dc25) (@vinhnx)
- Enhance token-efficient output handling and command parsing in executors (8ef8030c) (@vinhnx)
- Update queue display to show follow-ups and improve styling (cd8ee155) (@vinhnx)
- Remove unused fields and path handling from `HarnessEventEmitter` and `HarnessTurnState` (8321ad7f) (@vinhnx)
- Disable subagents by default and update documentation with usage and configuration details (acf6fb79) (@vinhnx)
- Transition `UnifiedCache` to use `RwLock` for interior mutability, enhance test coverage, and simplify cache operations (3aaf79c3) (@vinhnx)
- Remove `ui.show_message_dividers` config, simplify divider logic, and enhance tool summary rendering (eb57575e) (@vinhnx)
- Optimize memory usage and runtime efficiency, improve error handling, and enhance circuit breaker logic (79049f70) (@vinhnx)
- Remove `OpenAIPromptCacheSettings` import from `xai.rs` to clean up unused dependencies (670910ea) (@vinhnx)
- Remove `AnthropicProvider` and related implementations from the codebase to clean up unused functionality (16ac584b) (@vinhnx)
- Remove obsolete `src/acp/zed.rs` file and related references to streamline the codebase (123056a8) (@vinhnx)
- Remove Anthropic provider and OpenRouter implementation for codebase simplification (11636288) (@vinhnx)
- Remove Anthropic provider and OpenRouter implementation for codebase simplification (15e04793) (@vinhnx)
- Remove `models.rs` to simplify configuration and reduce redundancy in model management (fb4c67a5) (@vinhnx)
- Expand visibility for `parse_terminal_command` and `run_list_files` functions to improve module accessibility (c54b1e52) (@vinhnx)
- Reorder imports across modules for consistency and readability (3b9dcfce) (@vinhnx)
- Reorder imports across modules for consistency and readability (ce6524fa) (@vinhnx)
- Remove unused LLM request structures and related configurations (5a9a50e1) (@vinhnx)
- Remove `text_tools.rs` to simplify codebase and eliminate unused functions (31ed86b4) (@vinhnx)
- Remove `read_file_handler.rs`, `bash_runner.rs`, and unused code (839e9a8e) (@vinhnx)
- Remove `read_file_handler.rs`, `bash_runner.rs`, and unused code (2d7b2c6e) (@vinhnx)
- Adjust formatting, imports, and re-exports for improved consistency (80f3443c) (@vinhnx)
- Update tool policy to allow file editing and enhance output spooling for PTY-related tools (c4918c20) (@vinhnx)
- Update default model and tool policy permissions, add Codex harness learnings documentation (007d4c00) (@vinhnx)
- Switch default LLM provider to Anthropics, enhance reasoning deduplication, and apply Codex-inspired output limits (2ca1c343) (@vinhnx)
- Switch default provider to Hugging Face and add Moonshot Kimi K2.5 model support (5a956bb7) (@vinhnx)
- Switch default provider to Ollama, update model and API key configurations, and apply "if-let" refactoring for cleaner conditionals (dd0f2da4) (@vinhnx)
- Update default model to GPT-OSS, improve markdown rendering, and apply conditional/indentation optimizations (4e18aad8) (@vinhnx)
- Enhance line numbering in markdown code blocks, improve text trimming logic, and add support for "Reasoning" style rendering (fefc4e5b) (@vinhnx)
- Optimize markdown rendering with conditional line numbering, add diff language detection, and improve (dbf227de) (@vinhnx)
- Improve reasoning rendering logic, add deferred rendering support, and optimize duplicate content handling (21f794ef) (@vinhnx)
- Standardize color palette, update UI feedback styles, and improve markdown spacing configuration (e4aeee1b) (@vinhnx)
- Add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic (b310eabc) (@vinhnx)
- Add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic (e20c1227) (@vinhnx)
- Enhance rendering logic with dimming style, standardize long-running command locks, and update tooling policies for improved usability (a4e23c20) (@vinhnx)
- Enhance command status handling, improve loop detection logic, and update tool execution messages for clarity (98d776c1) (@vinhnx)
- Add follow-up prompts for truncated outputs and improve spooled file handling messages for enhanced user guidance (330af751) (@vinhnx)
- Update authorship information across multiple crates, enhance TUI performance with increased tick rates, and integrate new tui-shimmer dependency for improved UI effects (956147f6) (@vinhnx)
- Remove DESIGN_SYSTEM.md, update default model to minimax-m2.1:cloud, and improve TUI spinner handling with `is_spinner_frame` function (2193c976) (@vinhnx)
- Enhance rendering logic for consistent tool output styling, simplify message spans, and update spinner handling for better readability (bdbe29d1) (@vinhnx)
- Extract metric recording and remove auto-exit plan mode logic from tool execution result handling. (5e83a3a2) (@vinhnx)
- Centralize tool outcome handling parameters into a new `ToolOutcomeContext` struct. (73856006) (@vinhnx)
- Adjust tool outcome context passing and borrowing in turn processing. (a197876a) (@vinhnx)
- Refine `ToolOutcomeContext` lifetimes to improve mutable borrowing patterns and simplify context access. (a81254ef) (@vinhnx)
- Unify direct tool execution and expand interaction loop context with new tool-related services. (d7953097) (@vinhnx)
- Update tool execution methods and enhance context handling (e592f983) (@vinhnx)
- Replace println with tracing for improved logging consistency (4ad8b586) (@vinhnx)
- Replace eprintln with tracing for improved logging consistency (1d8f11d2) (@vinhnx)
- Update logging in TUI code to use tracing instead of println and eprintln (3393f2e0) (@vinhnx)
- Enhance local build process for macOS and Linux in release scripts (98cb01fb) (@vinhnx)
- Improve malloc warning suppression and enhance spinner behavior in UI interactions (241c53fe) (@vinhnx)
- Update malloc warning suppression in debug script (839dbd68) (@vinhnx)
- Enhance cursor visibility logic by adding status spinner check (910506f9) (@vinhnx)
- Implement color accessibility features and update configuration options (6422c103) (@vinhnx)
- Improve table formatting in color guidelines documentation (d33b7afa) (@vinhnx)
- Add Kimi K2.5 model support and update related configurations (6e83202d) (@vinhnx)
- Add Kimi K2.5 and GLM 4.7 models to Ollama provider documentation and tests (84b6c722) (@vinhnx)
- Improve formatting of model pull commands in Ollama provider documentation (252114fc) (@vinhnx)
- Update terminal theme to ciapre-dark (157b61cd) (@vinhnx)
- Reorder use statements for clarity in theme module (63495bea) (@vinhnx)
- Remove unused Wrap widget import from history picker (67d5cc2c) (@vinhnx)
- Add dead code allowance for search_position and start_search method (39dc545b) (@vinhnx)
- Improve code formatting and readability across multiple files (d4ba430d) (@vinhnx)
- Optimize development and test profiles in Cargo.toml (7f4cc397) (@vinhnx)
- Optimize development profile settings in Cargo.toml (cac65958) (@vinhnx)
- Update process group management documentation for clarity (aff7157a) (@vinhnx)
- Streamline tool call validation and enhance error handling (3b3055c3) (@vinhnx)
- Simplify command parsing and enhance diff formatting (57ddc381) (@vinhnx)
- Consolidate shared utilities and unify LLM streaming (02a78918) (@vinhnx)
- Unify response handling and fix reasoning details parsing (a342a531) (@vinhnx)
- Consolidate supported models into constants (2dcc9d6b) (@vinhnx)
- Streamline release script and remove unused error handling code (2a456008) (@vinhnx)
- Improve reasoning content comparison and suppress duplication in response rendering (87b066cf) (@vinhnx)
- Remove deprecated model constants and clean up supported models list (eb9b6ff9) (@vinhnx)
- Remove unused set_defer_rendering method from StreamingReasoningState (e429daf0) (@vinhnx)
- Update tool policies to allow all actions and improve terminal cleanup logic (0e54e2ff) (@vinhnx)
- Simplify conditional checks and remove unused imports (2602cbc8) (@vinhnx)
- Enhance base URL resolution and improve JSON handling in request builder (584a82ca) (@vinhnx)
- Reorganize release steps and update Homebrew process (e3d99f5a) (@vinhnx)
- Simplify workspace directory creation in first run setup (6a35774d) (@vinhnx)
- Remove scroll indicator from footer widget and associated UI logic. (4f060bed) (@vinhnx)
- Optimize I/O operations with buffered writes and simplify `ToolCallRecord`'s `tool_name` ownership. (a66a3a9d) (@vinhnx)
- Improve string truncation logic to ensure character boundaries are respected (83ebed35) (@vinhnx)
- Streamline release process and remove deprecated crate waiting logic (4436c3cc) (@vinhnx)
- Consolidate path resolution logic and remove redundant functions (d0a10bad) (@vinhnx)
- Optimize line truncation logic in summarizers (4996c6ad) (@vinhnx)
- Improve code readability and structure across multiple files (2b5a2895) (@vinhnx)
- Comprehensive optimization of agent loop and tool execution pipeline (e97e91c0) (@vinhnx)
- Optimize tool signature handling and caching; enhance turn configuration extraction and prompt caching (b42c331a) (@vinhnx)
- Add token tracking validation in ContextManager; optimize turn balancer check intervals (4dd087c9) (@vinhnx)
- Replace FxHashMap with LoopTracker for tool attempt tracking; optimize loop detection and history management (bb831f19) (@vinhnx)
- Improve code formatting and readability across multiple files (393f63e9) (@vinhnx)
- Remove unused imports in turn_loop.rs for cleaner code (936e1885) (@vinhnx)
- Enhance error handling for tool arguments and improve rate limiting logic (386ce6aa) (@vinhnx)
- Improve tool validation and error handling; enhance test coverage for non-interactive environments (4b47208b) (@vinhnx)
- Enhance IDE context flushing and user confirmation handling; improve command auditing (b00d3b7f) (@vinhnx)
- Streamline error handling and validation logic; enhance retry safety checks (754ce484) (@vinhnx)
- Update TODO with comprehensive code audit and optimization guidelines (4f96ff24) (@vinhnx)
- Replace hardcoded Plan Mode strings with constants for consistency and maintainability (5d4d2407) (@vinhnx)
- Remove dead code and streamline path handling functions (35289283) (@vinhnx)
- Streamline MCP event handling and enhance error content construction (9510ef79) (@vinhnx)
- Enhance tool validation and error messaging, modularize execution logic (2f6d20f0) (@vinhnx)
- Modularize tool output handling and enhance command safety validation (88146309) (@vinhnx)
- Enhance path validation logic and add lexical workspace check (605ea1ec) (@vinhnx)
- Enhance file output handling and add no_spool flag for read operations (a3c134c4) (@vinhnx)
- Increase spooling thresholds and improve output handling for large tool outputs (dbae38d2) (@vinhnx)
- Remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
- Reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- Extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- Streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- Consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- Streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- Remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- Clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- Remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- Make TUI signal cleanup and dotfile permission backup UNIX-specific. (14a4f2d2) (@vinhnx)
- Improve previous SemVer tag identification by searching commit history in release scripts (0fcdce3d) (@vinhnx)
- Remove unified tool resolver module to streamline codebase (1384fe03) (@vinhnx)
- Remove unused TUI components and improve event handling for focus changes (99fa2620) (@vinhnx)
- Standardize continuation prefix handling in wrap_block_lines functions (0596e899) (@vinhnx)
- Enhance changelog generation and update test script for versioning (4383577f) (@vinhnx)
- Update documentation and improve terminology for unified search and code understanding (57e584eb) (@vinhnx)
- Improve cacheable tool logic and enhance formatting in build script (2b276290) (@vinhnx)
- Streamline provider resolution and update configuration to use Minimax (2b1d09c7) (@vinhnx)
- Update Z.AI model constants and enhance parsing logic for legacy support (44e33c22) (@vinhnx)
- Remove OpenSSL setup for Windows and add Linux TLS linkage verification (5aac5ca2) (@vinhnx)
- Add support for Linux x86_64 musl binaries in build and release workflows (137992e8) (@vinhnx)
- Update CI workflows for Linux and Windows builds with optional artifact downloads (d72fb2d0) (@vinhnx)
- Enhance plan approval process with context clearing option and update confirmation modal (dd522f15) (@vinhnx)
- Simplify assignment of session_stats and improve formatting in model ID parsing (9b3273d6) (@vinhnx)
- Update LLM provider and default model in configuration (14b036a5) (@vinhnx)
- Streamline task tracking logic and introduce TaskTrackingStatus for improved status management (3be2c784) (@vinhnx)
- Rename sessions command to resume and update related handlers (c63ab822) (@vinhnx)
- Enhance panic handling, optimize agent harness, and improve token efficiency (c68e7ea1) (@vinhnx)
- Enhance error handling and recovery mechanisms; improve error categorization and logging (43a4e099) (@vinhnx)
- Remove unnecessary Clear widget calls in header and input rendering (fb1c8de7) (@vinhnx)
- Streamline architecture documentation and enhance clarity (565ea315) (@vinhnx)
- Rename color-eyre hook functions for clarity and consistency (3c555a86) (@vinhnx)
- Update approval policy handling to use Arc<RwLock> for thread safety (3b91595f) (@vinhnx)
- Create sync_mcp_approval_policy_for_context function for improved clarity (92a6cf82) (@vinhnx)
- Improve error handling and simplify string conversion in terminal color probe and LM Studio client (e41acb42) (@vinhnx)
- Update exit summary formatting and remove unused functions (7c58ff16) (@vinhnx)
- Streamline lint configuration and enhance exit summary output (eaca32a6) (@vinhnx)
- Enhance exit summary formatting and improve file indexing logic (d143fe97) (@vinhnx)
- Refactor: update terminology from "Human-Agent Teaming" to "Human-Agent Collaboration" in roadmap (af93444b) (@vinhnx)
- Update reference to plan mode and HITL flow structure (b159efb9) (@vinhnx)
- Streamline response handling and output printing in AgentRunner (d98a5e55) (@vinhnx)
- Simplify function output handling in HuggingFace and OpenAI providers (71381dd3) (@vinhnx)
- Enhance diff rendering styles and streamline background color handling (c85edbcf) (@vinhnx)
- Enhance color detection logic for terminal compatibility (af35cbf6) (@vinhnx)
- Update configuration section markers and adjust merging logic for user settings (79bce0e0) (@vinhnx)
- Enhance settings search functionality and improve nested key handling (81f799ca) (@vinhnx)
- Remove agent command and related logic, update slash command descriptions (d672e41a) (@vinhnx)
- Modularize components in src/ for improved readability and maintainability (f405950c) (@vinhnx)
- Enhance plugin ABI with safety checks and optional deallocator (ef02885a) (@vinhnx)
- Integrate unicode-width for text display handling (f7251f5f) (@vinhnx)
- Implement schema command for runtime tool introspection and add input validation (8d2900a1) (@vinhnx)
- Improve visual distinction between assistant agent and PTY output messages (387a45c7) (@vinhnx)
- Update PTY fallback handling for consistent terminal output (ed0fa6c9) (@vinhnx)
- Streamline ToolDefinition creation and improve serialization functions (5b3cc708) (@vinhnx)
- Streamline event processing and enhance modal state management (1e7c6031) (@vinhnx)
- Introduce `detect_pattern_with_guard` for improved pattern detection logic and simplify `RwLock` write guard acquisition. (5f70ca7c) (@vinhnx)
- Streamline OAuth handling and improve prompt rendering logic (91a3c73d) (@vinhnx)
- Enhance instruction rendering and highlight extraction in project documentation (cf116dac) (@vinhnx)
- Improve compaction handling and documentation across multiple files (fe86c4c5) (@vinhnx)
- Use get_config_dir for configuration directory retrieval in DotManager (28c82ef6) (@vinhnx)
- Replace provider resolution logic with infer_provider_from_model in create_provider_for_model (a8c208e3) (@vinhnx)
- Improve command processing and rendering efficiency in drive_terminal (e33ed10d) (@vinhnx)
- Update provider client creation logic and add debug logging for provider usage (9c947bde) (@vinhnx)
- Enhance conversation handling and user guidance in agent interactions (f16808a8) (@vinhnx)
- Enhance tool completion replies with LLM-generated suggestions and fallback logic (1ad0b3de) (@vinhnx)
- Integrate InlineHeaderContext into event handling and model selection (75c6bbc0) (@vinhnx)
- Consolidate session archive handling and improve debug log directory management (71f0b686) (@vinhnx)
- Reorganize imports and improve code formatting in interactive features (20817b85) (@vinhnx)
- Clean up code formatting and improve readability in multiple files (dffcbdae) (@vinhnx)
- Remove unused content delta handling in OpenAIStreamTelemetry (44da3f56) (@vinhnx)
- Remove PR review status handling and optimize TUI tick rates (737a2c75) (@vinhnx)
- Simplify error sanitization logic and preserve full error content (7a3afa95) (@vinhnx)
- Simplify model and provider selection entries, remove unnecessary enumerations (baea18fb) (@vinhnx)
- Streamline OpenAI session clearing by consolidating storage methods (23a6c55b) (@vinhnx)
- Improve git diff preview logic and update related enums (18a30a87) (@vinhnx)
- Update prompt assembly logic and enhance instruction handling (8b6f71f5) (@vinhnx)
- Remove unused PromptAssemblyMode and simplify related code (1f348cd4) (@vinhnx)
- Streamline system prompt instructions for clarity and conciseness (dcf8fa34) (@vinhnx)
- Enhance performance guidelines and simplify related documentation (f6259c32) (@vinhnx)
- Replace custom timing with /usr/bin/time for performance measurement (5be0c3b4) (@vinhnx)
- Update tool policies and improve command submission handling (890eacf0) (@vinhnx)
- Improve GitHub Copilot OAuth flow and update configuration settings (86b8d667) (@vinhnx)
- Remove explicit unsafe Send and Sync implementations for NativePlugin. (5c96c9d3) (@vinhnx)
- Simplify key handling in modal state (bfec2be3) (@vinhnx)
- Reuse shared OpenAI prompt cache key helper (68fe597f) (@vinhnx)
- Tighten system prompts and document accuracy tuning (450dcef2) (@vinhnx)
- Update tool policies and enhance prompt suggestions in TODO (36313870) (@vinhnx)
- Remove outdated prompt suggestion details and add subagent build task (c869b499) (@vinhnx)
- Enhance error handling and retry logic in tool execution (89075d3d) (@vinhnx)
- Improve error handling and classification in tool execution (4990c7dd) (@vinhnx)
- Enhance error handling and retry logic in agent execution (9f626e1e) (@vinhnx)
- Replace try_recv with recv for steering message handling (17edd05a) (@vinhnx)
- Introduce idle wake delay in inline event loop and related components (4fdec1dc) (@vinhnx)
- Streamline tool policy checks and improve error handling (25c11664) (@vinhnx)
- Replace RwLock with parking_lot::RwLock for pattern detector (278238fd) (@vinhnx)
- Add compact aliases for session_id and path parameters across tools (82340a2e) (@vinhnx)
- Enhance WebSocket handling for OpenAI provider, improve error detection, and update configuration documentation (9ae3e202) (@vinhnx)
- Enhance WebSocket handling and documentation for OpenAI provider (8df581c1) (@vinhnx)
- Streamline code structure and improve readability across multiple files (22d1fa22) (@vinhnx)
- Unify and streamline agent harness execution (cfcf3b05) (@vinhnx)
- Streamline error handling and formatting in tests and utils (55b6137b) (@vinhnx)
- Update parent death signal handling and improve FFI type safety (6995c383) (@vinhnx)
- Update context handling in interaction loop and improve test module organization (d74a4e56) (@vinhnx)
- Simplify function signature in test_execute_with_delay (4b26e988) (@vinhnx)
- Format dependencies and improve parent death signal handling (392f8fd5) (@vinhnx)
- Remove Vim mode command and related functionality (f3b44164) (@vinhnx)
- Simplify vtcode configuration (ab835726) (@vinhnx)
- Remove obsolete code (a55f265c) (@vinhnx)
- Remove contributors section from changelog template (30707f13) (@vinhnx)
- Remove FollowUp variant from ReplyKind and update related logic (6f12c2f7) (@vinhnx)
- Improve duration handling and validation in rate limiting and retry logic (3ec14baf) (@vinhnx)
- Unify tool execution kernel and align harness safety limits (5fbfc7a8) (@vinhnx)
- Clean up tool policies and improve CLI help output (b32129ff) (@vinhnx)
- Enhance scrolling behavior and selection adjustment in session management (34060deb) (@vinhnx)
- Convert async functions to return Future for improved performance (38fc4722) (@vinhnx)
- Replace manual string truncation with truncate_byte_budget utility for consistency (29a6dc34) (@vinhnx)
- Replace eprintln with tracing for consistent logging across modules (803235eb) (@vinhnx)
- Change function signatures to return &str instead of &'static str (8b315ada) (@vinhnx)
- Streamline tool policies and update reasoning effort defaults for OpenAI models (15e635af) (@vinhnx)
- Reorganize AGENTS.md for clarity and improved workflow guidance (4732382c) (@vinhnx)
- Streamline tool handling and validation logic, remove unused code (477319a0) (@vinhnx)
- Update skill manifest handling and permissions to use boxed types (aab2d60c) (@vinhnx)
- Replace direct file writes with write_private_file for secure permissions (0319c5e7) (@vinhnx)
- Reorganize imports and improve relaunch logic (91a841e1) (@vinhnx)
- Update installer scripts to fetch recent releases and improve error handling (1f45415f) (@vinhnx)
- Convert async functions to inline-delegating wrappers to reduce coroutine state machine overhead (5ea90914) (@vinhnx)
- Clean up unused dependencies and empty files via cargo-shear (a45d5063) (@vinhnx)
- Optimize atomic operations and session management in PTY handling (1141a096) (@vinhnx)
- Enhance retry logic with structured error handling and backoff steps (8552a810) (@vinhnx)
- Update PlaceholderSpinner initialization and add defer restore functionality (75660e9e) (@vinhnx)
- Clean up tool policies and adjust execution status handling (7564f2bf) (@vinhnx)
- Remove unused 'walkdir' dependency and clean up config watcher logic (514d5a3a) (@vinhnx)
- Streamline tool policies and enhance DeepSeek provider error handling (9f1a794f) (@vinhnx)
- Update message styles for tool output rendering (63263fa1) (@vinhnx)
- Simplify condition checks and enhance clause intent evaluation logic (8abcbc01) (@vinhnx)
- Update documentation links to use public methods in harness and pty facades (466825fc) (@vinhnx)
- Simplify status line layout handling and improve auto status updates (fe1c3b85) (@vinhnx)
- Optimize code for better performance and readability (577415b8) (@vinhnx)
- Optimize tool call counting and improve string handling in loop detector (88bd10aa) (@vinhnx)
- Optimize diff computation and improve whitespace normalization in file helpers (91413b28) (@vinhnx)
- Change floating-point types from f64 to f32 for consistency and performance (b350c7d0) (@vinhnx)
- Implement read_exact_uninit for optimized async reading in file operations (787cea68) (@vinhnx)
- Enhance stack protection by capping stack size in pre-main hardening functions (3530b905) (@vinhnx)
- Enhance credential encryption with per-file salt and improve URL validation in WebFetch (d49e58f6) (@vinhnx)
- Optimize loops for width calculations in table rendering (cdad5eff) (@vinhnx)
- Update windows-sys dependency to 0.61.2 and add pretty_assertions for enhanced testing (b6e98c07) (@vinhnx)
- Improve process hardening documentation and clarify stack limit handling (0d1c310c) (@vinhnx)
- Implement integer overflow handling and improve safety checks across multiple modules (e8da535c) (@vinhnx)
- Replace `format!` with `write!` and `writeln!` for improved performance and clarity in multiple files (341a2de9) (@vinhnx)
- Enhance integer overflow handling documentation and clarify performance implications (05a00f9a) (@vinhnx)
- Integrate mimalloc for improved memory management and optimize various data structures (cd05078d) (@vinhnx)
- Implement process hardening functions and streamline environment variable management (42e3528e) (@vinhnx)
- Add missing crate paths for vtcode dependencies (f2c4630f) (@vinhnx)
- Streamline model resolution and remove error handling duplication in OpenRouterProvider (764b088b) (@vinhnx)
- Remove MiMo V2 Pro and MiMo V2 Omni models from configurations and related files (6926f102) (@vinhnx)
- Streamline error handling and improve code readability across multiple files (966569c2) (@vinhnx)
- Update header context handling and improve reasoning effort configuration (bb71e501) (@vinhnx)
- Improve code formatting and readability in multiple files (a2716021) (@vinhnx)
- Route SHA256 hashing through vtcode-commons::utils::calculate_sha256 (0b73c956) (@vinhnx)
- Consolidate serde_json pretty-print fallback into vtcode-commons (500dca36) (@vinhnx)
- Regroup resilience primitives into tools::resilience (eea8d9aa) (@vinhnx)
- Improve code readability and consistency across multiple files (eb4b169c) (@vinhnx)
- Extract release-lib.sh, add cross-as-default with M4 tuning, optimize CI caching (9b035778) (@vinhnx)
- Replace cross with cargo-zigbuild for cross-compilation and update packaging scripts (d09699ed) (@vinhnx)
- Update build workflows for Linux and Windows binaries (792337eb) (@vinhnx)
- Improve line number styling and update reasoning effort to high (7398263f) (@vinhnx)
- Update README structure and improve formatting for clarity (4c9f5df1) (@vinhnx)
- Streamline README by removing outdated sections and enhancing clarity (303de2d6) (@vinhnx)
- Remove perg fallback and update grep_file tool documentation (703dfda2) (@vinhnx)
- Update tool output handling and improve user guidance for reading content (1eb305ba) (@vinhnx)
- Extract shared provider boilerplate and consolidate error types (5df85e50) (@vinhnx)
- Consolidate DRY violations across 4 crates (64ad80b2) (@vinhnx)
- Implement rule fill for Fieldset-style blocks and update tests (7fe7c4ad) (@vinhnx)
- Rename spinner frame function and update references (edc0c058) (@vinhnx)
- Improve file selection logic and scoring in file palette (8bd71623) (@vinhnx)
- Improve table rendering and wrapping logic (c67f9d26) (@vinhnx)
- Optimize table rendering logic and add space padding function (b56f3551) (@vinhnx)
- Remove vt100 legacy PTY backend, keep GhosttyCore only (4bd68ffe) (@vinhnx)
- Update README layout and improve visual presentation (4aa9ef77) (@vinhnx)
- Update README layout and improve feature descriptions (04cd5e54) (@vinhnx)
- Reorganize contributing section and update contributor images in README (10ec98b1) (@vinhnx)
- Move contributing section to the end of README and update contributor images (61b61d68) (@vinhnx)
- Merge vtcode-design, vtcode-theme, vtcode-tui into vtcode-ui (1f13c2fa) (@vinhnx)
- Remove TODO.md entry from documentation map (72a0f9c6) (@vinhnx)
- Format code for improved readability and consistency (0dcbbaaa) (@vinhnx)
- Consolidate vtcode-tui and vtcode-design into vtcode-ui, update documentation and references (c8d1cf29) (@vinhnx)
- Simplify session_archive usage in session loop runner (1a7ccf90) (@vinhnx)
- Streamline tool execution success handling and error reporting (69ee6119) (@vinhnx)
- Remove obsolete core agent loop exploration plans and tasks (3edf5e1d) (@vinhnx)
- Move float_to_json_number to common module and update usages (98032959) (@vinhnx)
- Simplify string formatting in tests and error handling (802e06ad) (@vinhnx)
- Simplify archived prompts loading and adjust reasoning effort level (c6966b04) (@vinhnx)
- Remove deprecated Qwen models from openrouter_models.json (d0c22c32) (@vinhnx)
- Normalize legacy tool names in permission rules and enhance warnings (ecab9d29) (@vinhnx)
- Remove Claude Fable 5 references and update to Claude Opus 4.8 across models and capabilities (3d0cd491) (@vinhnx)
- Enhance PTY session output handling and add utility functions (71951c99)
#### Tests
- Cover codex prompt filtering (37815eb0) (@vinhnx)
- Add missing fields to LLMRequest initializers (1040c352) (@vinhnx)
- Add streaming event deserialization tests (04c355ec) (@vinhnx)
- Add environment variable handling and cleanup in `test_get_gemini_api_key_from_config` (007bbd9a) (@vinhnx)
- Skip TUI-dependent tests in non-interactive environments (4559d31e) (@vinhnx)
- Add unit test for markdown table header rendering (db088ed7) (@vinhnx)
- Allow unsafe code in create_env_preserves_non_utf8_path test (524a1518) (@vinhnx)
- Enhance nested directories test to filter file matches (93d8e397) (@vinhnx)
- Add regression guard for vtcode_core::safety::hitl reachability (462b4e61) (@vinhnx)
## 0.128.4 - 2026-06-13
### Highlights
### Other Changes
## 0.128.3 - 2026-06-13
### Highlights
### Other Changes
#### Other
- Update TODO (b0a2c9f3)
- Add MiMo authentication method support and related configurations (b23ac5d1)
- Update TODO (ddf079ae)
- Add MiMo authentication method support and update UI constants (f80f806d)
## 0.128.2 - 2026-06-13
### Highlights
### Other Changes
## 0.128.0 - 2026-06-13
### Highlights
#### Features
- Feat/refactor OpenAI chatgpt Rig (8a57c899) (@kernitus)
- Add compile-time validation for tool name constants and integrate ctor for pre-main checks (394677b2)
- Add Kimi K2.7 Code model support across configurations and descriptions (1767daec)
### Other Changes
#### Other
- Update TODO (c272494b) (@vinhnx)
#### Refactors
- Normalize legacy tool names in permission rules and enhance warnings (ecab9d29)
- Remove Claude Fable 5 references and update to Claude Opus 4.8 across models and capabilities (3d0cd491)
## 0.127.0 - 2026-06-12
### Highlights
### Other Changes
#### Refactors
- Remove deprecated Qwen models from openrouter_models.json (d0c22c32)
## 0.126.0 - 2026-06-12
### Highlights
#### Bug Fixes
- Pin ratatui-widgets to =0.3.0 to fix cargo install (c60e1663) (@vinhnx)
- Fallback to unauthenticated GitHub API on 401 (fdc666f9)
- Handle terminal clear failure during force redraw (b85bec6e)
#### Documentation
- Remove Homebrew installation instruction and update provider link text (db9e269c) (@vinhnx)
- Update README to replace tables with bullet points for features and documentation (2e9081a3) (@vinhnx)
#### Features
- Add Kimi K2.7 Code model and update related configurations (5f74a144)
- Implement double-Esc detection for rewind functionality (ca7a521b)
- Add MiniMax-M3 model via Novita inference provider and update related configurations (ecdd32f4)
### Other Changes
#### Other
- Update TODO (d54a296f) (@vinhnx)
- Update README (27a96ce8) (@vinhnx)
- Add release notes command and display on startup after updates (8f6848fd)
- Refactor model provider mappings and remove deprecated models (9ead7f9c)
#### Refactors
- Move float_to_json_number to common module and update usages (98032959) (@vinhnx)
- Simplify string formatting in tests and error handling (802e06ad) (@vinhnx)
- Simplify archived prompts loading and adjust reasoning effort level (c6966b04)
## 0.125.3 - 2026-06-11
### Highlights
### Other Changes
#### Other
- Fix/671 chatgpt OpenAI requests (071c9e64) (@kernitus)
## 0.125.2 - 2026-06-11
### Highlights
### Other Changes
## 0.125.1 - 2026-06-11
### Highlights
#### Bug Fixes
- Use token-based auth for homebrew-tap push instead of gh credential (66ac78b3) (@vinhnx)
### Other Changes
#### Other
- Update TODO (b7d216f1) (@vinhnx)
- Fable (8781f2a4) (@vinhnx)
- Update README (7565cbac) (@vinhnx)
- Update README (7f4186dd) (@vinhnx)
- Ignore vtcode plan artifact (20736d87) (@vinhnx)
- Feat/merge modes & agents (3e743390) (@kernitus)
#### Refactors
- Streamline tool execution success handling and error reporting (69ee6119) (@vinhnx)
- Remove obsolete core agent loop exploration plans and tasks (3edf5e1d) (@vinhnx)
## 0.125.0 - 2026-06-10
### Highlights
#### Features
- Add vtcode-ui to publish sequence and update CLI documentation (bb7228ed)
### Other Changes
## 0.124.0 - 2026-06-10
### Highlights
#### Bug Fixes
- Handle parameterised close tags in collect_pseudo_marker_regions (dac7afa0)
- Apply PR review fixes (d530f6c7)
- Restore harness robustness plan section in TODO.md (84a652ea)
- Update tool call region collection to use pseudo marker (c67d347c)
- Improve screenshot path resolver to stop consuming trailing text (2b101bb1)
- Improve screenshot path resolver and extract shared path utilities (75e445c4)
- Update path handling in trim_trailing_image_path_str function (59208038)
- Add image handling in request processing and refactor fallback logic (52f780ae)
- Pin ratatui to =0.30.0 to prevent CellEffect Send breakage (8539a774)
#### Documentation
- Enhance ast-grep documentation in AGENTS.md and TODO.md; update system prompt to prefer ast-grep for code-shape queries (50afb69a)
#### Features
- Implement strip_textual_tool_call_regions function and update related error messages (c62e4253)
- Enhance recovery handling for tool-call markers and improve documentation (e30b9e68)
- Add Claude Fable 5 with fallback credit, remove deprecated models (d4ed7872)
- Enhance Anthropic provider with new request handling and response parsing (73002a49)
- Add support for CLAUDE_OPUS_4_8 in thinking profile and update test case (24fc08e8)
### Other Changes
#### Other
- Remove outdated documentation references and clean up project files (ebdfe55f)
- Update ignore (114a62ec)
#### Refactors
- Merge vtcode-design, vtcode-theme, vtcode-tui into vtcode-ui (1f13c2fa)
- Remove TODO.md entry from documentation map (72a0f9c6)
- Format code for improved readability and consistency (0dcbbaaa)
- Consolidate vtcode-tui and vtcode-design into vtcode-ui, update documentation and references (c8d1cf29)
- Simplify session_archive usage in session loop runner (1a7ccf90)
## 0.123.10 - 2026-06-09
### Highlights
### Other Changes
## 0.123.9 - 2026-06-09
### Highlights
### Other Changes
## 0.123.8 - 2026-06-09
### Highlights
#### Documentation
- Enhance agent and subagent documentation with primary agent details (4cb84e23)
- Add screenshot spotlight section with GIF and image table (2b2252a0)
#### Features
- Implement subagent lifecycle tool policy and enhance subagent controller notifications (017ca5bf)
- Add support for new package managers and improve version handling (9dda3341)
### Other Changes
#### CI
- Improve Rust toolchain management with MSRV check and nightly smoke test (9909ad4b)
#### Other
- Update TODOs (c47dd4fb)
- Update TODO (d41ae5ed)
- Refactor tool registration to use distributed slices and update README contributors script (1f1c6978)
#### Refactors
- Update README layout and improve visual presentation (4aa9ef77)
- Update README layout and improve feature descriptions (04cd5e54)
- Reorganize contributing section and update contributor images in README (10ec98b1)
- Move contributing section to the end of README and update contributor images (61b61d68)
## 0.123.7 - 2026-06-08
### Highlights
### Other Changes
## 0.123.6 - 2026-06-08
### Highlights
#### Bug Fixes
- Resolve docs.rs build failure for vtcode-config (5f722204)
#### Features
- Enrich ast-grep structural search with metaVariables, byteOffset, and url fields (5dcbc31f)
### Other Changes
#### Other
- Feat/top level agents (533af16b) (@kernitus)
- Refactor file exclusion logic and enhance directory walking utilities (21119580)
- Enhance documentation and configuration for GhosttyCore integration; update related tests and clean up lint settings (0d210b1e)
- Update tree-sitter grammar handling and enhance unified search parameters for rewrite workflows (4eaa8c22)
- Add linting rules for Kotlin, Python, Ruby, and Rust to enforce best practices (f44d5691)
- Add linting rules for Python, Rust, TypeScript, and TSX to enforce best practices (e049bc14)
- Refactor and enhance linting rules for Python and Kotlin; add new rules for identity checks and update metadata for existing rules (acb69e45)
- Add rule snapshots for various code quality improvements (4ba0b8dd)
- Add tests for relational rule YAML generation and validation (85c1f44f)
- Add support for C and C++ in local parser checks; update tests accordingly (e51eceee)
- Refactor StructuralSearchRequest to use serde_json types directly; remove unused validation functions (cbb3d734)
- Add new rules and updates for various languages; enhance JSON schema for unified search parameters (db75ccf4)
- Remove unused dependencies, enhance documentation, and update various rules for improved code quality and consistency across languages. (3cd68c2a)
- Remove stale lint expectations that are now fulfilled (9aa04c7d)
#### Refactors
- Remove vt100 legacy PTY backend, keep GhosttyCore only (4bd68ffe)
## 0.123.5 - 2026-06-07
### Highlights
#### Bug Fixes
- Resolve 3 critical harness robustness bugs from session turn_364 (ebdbafb1)
#### Documentation
- Add AGENTS.md files for various VT Code components (d56ab044)
#### Features
- Add DRY CLI messages helper and detailed clap help for agents (f37e27c8)
- Enhance tool permission handling and improve ANSI escape sequence detection (b6e069ce)
### Other Changes
#### Other
- Remove Ghostty VT runtime integration and related tests; adjust terminal size in tests. (2218ca09)
- Update TODO (8f62b657)
## 0.123.4 - 2026-06-07
### Highlights
### Other Changes
## 0.123.3 - 2026-06-06
### Highlights
#### Bug Fixes
- Add vtcode-macros to publish order before vtcode-core (55f2e4df) (@vinhnx)
- Make crate publishing non-fatal in release.sh and add dependency validation to publish script (12ef4e7b) (@vinhnx)
- Remove double escape functionality and update related hints (4884c153)
- Suppress deprecation warnings for anstyle utilities (27aa88f9)
### Other Changes
#### Other
- Refactor file operations and enhance error handling (e09bf2b5)
- Add unified style bridging between anstyle and ratatui (47e14082)
## 0.123.2 - 2026-06-06
### Highlights
#### Bug Fixes
- Add vtcode-ghostty-core to publish order before vtcode-core (eeb15246)
### Other Changes
## 0.123.1 - 2026-06-06
### Highlights
#### Bug Fixes
- Resolve double escape functionality issue and improve response handling (da06eb28)
- Skip OS keyring during tests/CI to avoid Keychain prompts (a60f1721)
- Ensure deterministic hashing by sorting JSON object keys (38fe3afb)
#### Documentation
- Enhance README with detailed command usage and workflows (62b2a576)
- Add sponsorship section with links and current sponsors (46b40322)
- Update feature and provider tables for clarity and consistency (1bc46f46)
- Add allocation optimization plan to eliminate redundant heap allocations (acc0a4ba)
#### Features
- Add support for Nemotron 3 Ultra and Gemma 4 models with updated configurations (16e6026d)
### Other Changes
#### Other
- Audit and refactor (9a26456c) (@vinhnx)
- Update API key and image sizes in README.md (b7c60aa6) (@vinhnx)
- Simplify quick start instructions for VT Code (3dff646f) (@vinhnx)
- Add vtcode update command to README (974cfd25) (@vinhnx)
- Enhance images in README with rounded corners; update TODO and configuration settings (87a81d8d)
- Refactor tool name handling to use CompactStr for improved memory efficiency (8c1c7a30)
- Add openrouter code generation and macros for string newtypes (b875258a)
- Implement core terminal functionality with screen management, input processing, and device reporting (ae290963)
#### Refactors
- Improve file selection logic and scoring in file palette (8bd71623)
- Improve table rendering and wrapping logic (c67f9d26)
- Optimize table rendering logic and add space padding function (b56f3551)
## 0.123.0 - 2026-06-05
### Highlights
#### Features
- Add NVIDIA Nemotron 3 Ultra model and update provider logic (b27b6cbf)
- Add NVIDIA Nemotron 3 Ultra model and update related configurations (bff99dc1)
- Enhance JSON parsing to recover from truncated input (989d3bba)
### Other Changes
## 0.122.0 - 2026-06-05
### Highlights
#### Features
- Add Evolink provider (OpenAI-compatible multi-model gateway) (bdc4c753) (@EvoLinkAI)
- Update file palette tree defaults and styling (3af4701b)
- Add Evolink provider support and update related documentation (e9577426)
- Update OpenAI function parameter sanitization and change default model in configuration (8871c397)
- Add new Evolink models and update configurations (8aafefb9)
### Other Changes
#### Other
- Update TODO (017be561) (@vinhnx)
- Add help modal using ratatui-cheese for keyboard shortcuts (87f3630e) (@vinhnx)
- Update UI highlight symbols and styles across multiple components; change theme to "vitesse-black". (41503473) (@vinhnx)
- Refactor input handling and styles across various components; unify search field rendering; update theme to ciapre-blue. (33a910d6)
#### Refactors
- Extract shared provider boilerplate and consolidate error types (5df85e50) (@vinhnx)
- Consolidate DRY violations across 4 crates (64ad80b2) (@vinhnx)
- Implement rule fill for Fieldset-style blocks and update tests (7fe7c4ad) (@vinhnx)
- Rename spinner frame function and update references (edc0c058) (@vinhnx)
## 0.121.1 - 2026-06-03
### Highlights
### Other Changes
#### CI
- Fix x86_64-unknown-linux-gnu aws-lc-sys gcc-9 panic (58a30561) (@vinhnx)
- Enable Windows builds by default in release workflows (7bce5b8d) (@vinhnx)
#### Refactors
- Improve line number styling and update reasoning effort to high (7398263f) (@vinhnx)
- Update README structure and improve formatting for clarity (4c9f5df1)
- Streamline README by removing outdated sections and enhancing clarity (303de2d6)
- Remove perg fallback and update grep_file tool documentation (703dfda2)
- Update tool output handling and improve user guidance for reading content (1eb305ba)
## 0.121.0 - 2026-06-03
### Highlights
### Other Changes
## 0.120.0 - 2026-06-03
### Highlights
### Other Changes
#### Refactors
- Update build workflows for Linux and Windows binaries (792337eb)
## 0.119.0 - 2026-06-03
### Highlights
### Other Changes
## 0.118.0 - 2026-06-03
### Highlights
#### Features
- Add Xiaomi MiMo V2.5 and V2.5 Pro models with updated documentation and default settings (608b7fca)
### Other Changes
#### Other
- Update README (f2b4202b)
## 0.117.7 - 2026-06-03
### Highlights
#### Bug Fixes
- Skip Windows builds when targets missing, upload assets individually (ced93f5f)
### Other Changes
## 0.117.6 - 2026-06-03
### Highlights
#### Bug Fixes
- Upload failure, homebrew update, skip aarch64 cross on mac (44ea5199) (@vinhnx)
### Other Changes
#### Refactors
- Replace cross with cargo-zigbuild for cross-compilation and update packaging scripts (d09699ed)
## 0.117.5 - 2026-06-02
### Highlights
#### Bug Fixes
- Run cross builds sequentially to avoid OOM on 16GB M4 (5ad249b4)
### Other Changes
## 0.117.4 - 2026-06-02
### Highlights
#### Bug Fixes
- Cross-build platform compat, yanked core2 dep (a935bace)
- Build windows arm64 natively, cross images require ghcr auth (395c16d5)
### Other Changes
## 0.117.3 - 2026-06-02
### Highlights
### Other Changes
## 0.117.2 - 2026-06-02
### Highlights
#### Bug Fixes
- Pin compatible protocol SDK (ec6399d2) (@vinhnx)
- Update API key and default model settings in vtcode.toml (fc4c1dd6)
#### Documentation
- Add validated model pool guidance (278975c1) (@lucaszhu-hue)
- Document calculate_sha256 return format and use cases (c17ecad0) (@vinhnx)
- Update contributing guide for AI and human contributors (bca21a97)
#### Features
- Add MiniMax-M3 cloud model and update configurations (4d8b2cd2) (@vinhnx)
- Feat(paths): add expand_tilde function for home directory expansion (f92ab297) (@vinhnx)
- Add support for aarch64 target in Linux and update Homebrew formula (61f9f979)
### Other Changes
#### Refactors
- Route SHA256 hashing through vtcode-commons::utils::calculate_sha256 (0b73c956) (@vinhnx)
- Consolidate serde_json pretty-print fallback into vtcode-commons (500dca36) (@vinhnx)
- Regroup resilience primitives into tools::resilience (eea8d9aa) (@vinhnx)
- Improve code readability and consistency across multiple files (eb4b169c)
- Extract release-lib.sh, add cross-as-default with M4 tuning, optimize CI caching (9b035778)
#### Tests
- Add regression guard for vtcode_core::safety::hitl reachability (462b4e61) (@vinhnx)
## 0.117.1 - 2026-06-01
### Highlights
#### Features
- Add help modal for keyboard shortcuts and update related tests (20b360a1)
### Other Changes
#### Other
- Fix installer filename formatting (86e9fcc7) (@uiYzzi)
## 0.117.0 - 2026-06-01
### Highlights
#### Documentation
- Update README with Xiaomi MiMo V2.5 promotion details and add promotional image (786fa660)
#### Features
- Add MiniMax-M3 model with multimodal capabilities and update default model (64b27402)
### Other Changes
## 0.116.4 - 2026-05-31
### Highlights
#### Features
- Add launch instruction after successful update installation (ce65af75)
- Implement Readline-style keybindings for input editing and navigation (b5572979)
- Enhance modal key handling and update quick help lines (08a09dad)
### Other Changes
#### Other
- Update dependencies and enhance input management with undo/redo functionality (1b932528)
- Refactor LLM provider implementations to remove backend_kind method and improve code readability (fc12014f)
## 0.116.3 - 2026-05-31
### Highlights
#### Features
- Add rebindable keybinding system with config integration (ddc2d952)
- Implement environment variable for non-interactive trust management (ed395835)
- Add workspace trust management details to documentation (a135f12b)
- Enable auto-trust for workspace during non-interactive execution (895437d8)
- Update default model and increase max tool calls per turn (7f6c3414)
- Add new models for DeepInfra and Novita inference providers (65527ea4)
- Add support for launching external editor with draft text (932f7197)
- Add Next.js agent evaluation details and leaderboard status (7f2faf53)
- Update README for improved clarity and structure (bb4ee130)
- Enhance badge section in README for better visibility and organization (30515213)
- Update README for improved clarity and organization (33c4a589)
- Add release packaging automation and shell completions support (575e4449)
### Other Changes
#### Other
- Update TODO (2cb0c40b)
#### Refactors
- Improve code formatting and readability in multiple files (a2716021)
## 0.116.2 - 2026-05-31
### Highlights
#### Bug Fixes
- Update provider checks to include additional model names (52a57555)
### Other Changes
#### Other
- Fix dependency (208d6396) (@vinhnx)
## 0.116.1 - 2026-05-31
### Highlights
#### Features
- Add start and stop event stream commands to InlineCommand (2897d600)
### Other Changes
## 0.116.0 - 2026-05-30
### Highlights
### Other Changes
#### Other
- Update TODO (c4e2d4a0) (@vinhnx)
- Update changelog (9ba9979b) (@vinhnx)
- Add support section with donation link in README (cfd63e1d) (@vinhnx)
- Refactor LLM provider handling and improve error context management (b2cd624a)
- Remove deprecated DeepSeek V3.2 model references and update related tests (5ce314cf)
- Refactor model IDs and descriptions; remove deprecated models and update defaults (9b967eeb)
- Add LlamaCpp provider support and related configurations (12c5982c)
- Add support for managing local inference servers and new model integration (82ae1a5c)
- Add Ctrl+M shortcut to open model picker and update related tests (73538876)
- Add donation command and update related UI prompts (fd1a354d)
- Fix minimum Ollama version for Responses API and update model fields to be optional (4dc547e3)
- Add PR code review agent and related instructions (4f443419)
- Add PR code review guidelines to AGENTS.md (14d0e5cc)
- Add workspace structure and build guidelines to AGENTS.md (5009acb7)
- Refactor AGENTS.md to streamline rules and workspace structure (0ad20837)
- Add AGENTS.md files for new modules and update .gitignore to include memory directory (2be6f557)
- Update LM Studio provider to support new models and enhance API integration (86671657)
- Refactor file operations to use async Tokio methods for improved performance (e4582be4)
- Add criterion for benchmarking and improve text editing functions (dc22c528)
## 0.115.0 - 2026-05-30
### Highlights
#### Features
- Enhance deserialization for optional strings and update PlannerResponse structure (3d6643d2)
### Other Changes
#### Other
- Add StepFun provider and model integration (c3f73b10)
#### Refactors
- Update header context handling and improve reasoning effort configuration (bb71e501)
## 0.114.0 - 2026-05-29
### Highlights
#### Bug Fixes
- Remove setrlimit/getrlimit wrappers to fix type mismatch on Linux (eb2744d1)
- Update validation messages for structured outputs and task budget tokens in Claude Opus 4.8 (705f7def)
- Update API response thinking modes and configuration defaults for Anthropic provider (b814f5f5)
- Expose matches_model function for public use and update validation logic (4da619bf)
- Update model descriptions for Claude Haiku and Claude Sonnet to enhance clarity and detail (ee381fc1)
#### Documentation
- Update LLM provider integration instructions for clarity (a3f6166b)
#### Features
- Add support for mid-conversation system messages in Claude Opus 4.8 (076d8162)
- Add Laguna XS.2 model support and update configurations (5782eb5b)
### Other Changes
#### Other
- Update Anthropic model to Claude Opus 4.8 and adjust related configurations (6ec1aecf)
- Add Poolside provider and models support (4b9aa40d)
- Add compaction support to response structures and documentation (4ad4f344)
#### Refactors
- Remove MiMo V2 Pro and MiMo V2 Omni models from configurations and related files (6926f102)
- Streamline error handling and improve code readability across multiple files (966569c2)
## 0.113.0 - 2026-05-28
### Highlights
#### Bug Fixes
- Update parameter types for setrlimit and getrlimit functions (700440e8)
#### Features
- Add HTTP headers for OpenRouter requests (66d456a9)
### Other Changes
#### Other
- Add support for Qwen models in the model ID system (7445df05)
#### Refactors
- Streamline model resolution and remove error handling duplication in OpenRouterProvider (764b088b)
## 0.112.0 - 2026-05-28
### Highlights
#### Bug Fixes
- Remove invalid -C probe-stack codegen flag (74cd4eda) (@vinhnx)
### Other Changes
#### Other
- Add Xiaomi MiMo models and provider integration (fa121dd7)
## 0.111.1 - 2026-05-28
### Highlights
### Other Changes
#### Other
- Fix release script (6636c6bc)
## 0.111.0 - 2026-05-28
### Highlights
### Other Changes
## 0.110.0 - 2026-05-28
### Highlights
### Other Changes
## 0.109.0 - 2026-05-27
### Highlights
#### Bug Fixes
- Record approval decisions for pattern learning in handle_builtin_permission (39cc8183)
#### Features
- Update default model and add support for codex provider (1f361157)
### Other Changes
#### Other
- Add Mistral AI models and provider integration (81051c7a)
#### Refactors
- Add missing crate paths for vtcode dependencies (f2c4630f)
## 0.108.4 - 2026-05-27
### Highlights
#### Features
- Add DeepSeek balance retrieval and display functionality (9c6aacc8)
- Enhance navigation loop detection and add DeepSeek balance tracking (5e9bdf89)
- Update balance tracking for multiple providers and refactor related logic (d3739bd3)
- Improve Linux installation logic and enhance balance fetching timeout handling (8a51b9a0)
### Other Changes
#### Other
- Update TODO (f679482a)
- Update vtcode (9aa12ed8)
- Refactor enums to derive Copy trait for improved performance and memory efficiency (bf7f239b)
- Optimize performance and memory usage across various modules by utilizing `Vec::with_capacity`, `entry` API, and reducing unnecessary clones. Mark cold paths with `#[cold]` for better inlining control. Refactor string manipulations for efficiency and clarity. (3950f5f7)
- Refactor formatting functions to use `vtcode_commons::formatting::collapse_whitespace` for whitespace handling and optimize string concatenation using `write!` for improved performance. (0273970c)
#### Refactors
- Update documentation links to use public methods in harness and pty facades (466825fc)
- Simplify status line layout handling and improve auto status updates (fe1c3b85)
- Optimize code for better performance and readability (577415b8)
- Optimize tool call counting and improve string handling in loop detector (88bd10aa)
- Optimize diff computation and improve whitespace normalization in file helpers (91413b28)
- Change floating-point types from f64 to f32 for consistency and performance (b350c7d0)
- Implement read_exact_uninit for optimized async reading in file operations (787cea68)
- Enhance stack protection by capping stack size in pre-main hardening functions (3530b905)
- Enhance credential encryption with per-file salt and improve URL validation in WebFetch (d49e58f6)
- Optimize loops for width calculations in table rendering (cdad5eff)
- Update windows-sys dependency to 0.61.2 and add pretty_assertions for enhanced testing (b6e98c07)
- Improve process hardening documentation and clarify stack limit handling (0d1c310c)
- Implement integer overflow handling and improve safety checks across multiple modules (e8da535c)
- Replace `format!` with `write!` and `writeln!` for improved performance and clarity in multiple files (341a2de9)
- Enhance integer overflow handling documentation and clarify performance implications (05a00f9a)
- Integrate mimalloc for improved memory management and optimize various data structures (cd05078d)
- Implement process hardening functions and streamline environment variable management (42e3528e)
## 0.108.3 - 2026-05-24
### Highlights
#### Bug Fixes
- Update stop hook to use final text for improved response handling (24739e82)
- Ensure blank lines are rendered as placeholders in diff output (2395d8ea)
#### Features
- Enhance continuation evaluation with relaxed intent handling and add inline preview for spooled outputs (0d08141a)
### Other Changes
#### Other
- Update config (e95a9d64)
## 0.108.2 - 2026-05-24
### Highlights
#### Bug Fixes
- Improve tool output rendering for command success with no output (de2f5bad)
- Ensure duration calculations do not exceed maximum limits (c0fecf23)
#### Features
- Add 'file' alias for file path parameters across tools (381be8e5)
- Update tool policies and enhance DeepSeek provider handling (afaad984)
- Add new tools and update policies in tool-policy.json; refactor diff rendering to use standard ANSI colors (991d1590)
- Enhance tool output handling and add command summary recording (66d2691d)
- Add DSML parsing functionality and update reasoning effort configuration (fd749d7c)
- Enhance tool policy management and improve recovery handling (e4ef7275)
- Expose RecoveryPhase enum and improve recovery contract violation handling (30bde025)
- Implement DSML markup stripping and enhance navigation loop guidance (9e6b10a6)
- Remove tool policy JSON and enhance auto-approval logic in permission prompts (bfea75d4)
- Add hide header option for TUI to enhance user interface customization (b73d94fb)
- Improve tool policy permission management and remove auto-scroll functionality (0b833ca9)
- Enhance interim text continuation logic and improve conclusive marker detection (b76b7675)
- Refactor interim text continuation logic and enhance conclusive marker detection (2e5b96ea)
- Add command word splitting functionality for shell approval commands (3467ab3d)
- Optimize vector initializations and add #[must_use] annotations for clarity (e90ee59b)
- Update user message handling to use string slice for efficiency (854f732f)
### Other Changes
#### Other
- Update config (632d5235)
- Refactor completion handling and enhance tool recovery logic (ce916001)
- Optimize string allocations and improve performance in various modules (2300d49f)
- Refactor environment variable handling with env_lock; remove unsafe code in tests (407414e0)
- Update tool policies to allow additional actions and modify reasoning effort handling (2266862a)
#### Refactors
- Streamline tool policies and enhance DeepSeek provider error handling (9f1a794f)
- Update message styles for tool output rendering (63263fa1)
- Simplify condition checks and enhance clause intent evaluation logic (8abcbc01)
## 0.108.1 - 2026-05-22
### Highlights
### Other Changes
#### Other
- Update brew (36328c56)
## 0.108.0 - 2026-05-22
### Highlights
#### Bug Fixes
- Enable atlas cloud custom provider flows (84750086) (@lucaszhu-hue)
#### Features
- Update model configs and wire models list through custom provider (76f5146c) (@vinhnx)
### Other Changes
#### Other
- Refactor tool outcome handlers and improve recovery guidance (1af72c93)
- Refactor token usage tracking and simplify recovery logic in turn loop (a284faf9)
## scode-v0.50.14 - 2026-05-21
### Highlights
#### Features
- Add DeepSeek V4 Flash and Pro models to Hugging Face integration (f7aac344) (@vinhnx)
- Implement auto-scroll toggle functionality and update related key bindings (b9e91ef5) (@vinhnx)
### Other Changes
#### Other
- Remove deprecated OpenRouter models and update data structures for better type handling (06ccb60f) (@vinhnx)
- Refactor AgentRunner initialization and remove unused TUI session files (7da13a76) (@vinhnx)
- Remove integration example and MCP chat adapter files; update backend integration with ChatMessage interface (31e6dafc) (@vinhnx)
#### Refactors
- Enhance retry logic with structured error handling and backoff steps (8552a810) (@vinhnx)
- Update PlaceholderSpinner initialization and add defer restore functionality (75660e9e) (@vinhnx)
- Clean up tool policies and adjust execution status handling (7564f2bf) (@vinhnx)
- Remove unused 'walkdir' dependency and clean up config watcher logic (514d5a3a) (@vinhnx)
## 0.106.0 - 2026-05-20
### Highlights
### Other Changes
#### Other
- Implement slash command handling and associated models; add tests for command parsing and outcomes (2d07c12f)
- Add status line and terminal title setup commands (edca2bfb)
- Add support for inline loop actions and enhance diagnostics commands (34fc9261)
- Decompose TUI runloop and slash-command modules into smaller units (02d94320)
- Add interview context and payload handling for Planning workflow (d0516994)
- Add recovery preview and response handling for assistant interactions (d0b6e8e8)
- Refactor canonical_tool_name to return &str and update related usages (0dbef927)
- Refactor agent_type handling to use String instead of Cow and optimize message content processing (3901951c)
- Refactor code for readability and maintainability; streamline tool loop limit checks and enhance plan response handling (3d222d8e)
- Enhance Clippy lints for better safety and correctness checks (8b03a3e1)
- Replace #[allow(dead_code)] with #[expect(dead_code)] in multiple files to improve dead code handling (b01fb753)
- Add new tools to tool policy and adjust policies; update Clippy lints and test expectations (f9c6c5df)
- Update lints and improve code quality across multiple modules (ca987323)
- Add Gemini 3.5 Flash model (0509cebf)
#### Refactors
- Optimize atomic operations and session management in PTY handling (1141a096)
## 0.105.13 - 2026-05-14
### Highlights
### Other Changes
## 0.105.12 - 2026-05-14
### Highlights
### Other Changes
#### Other
- Update document about lacking support of Windows build (1757d58d) (@vinhnx)
#### Refactors
- Clean up unused dependencies and empty files via cargo-shear (a45d5063) (@vinhnx)
## 0.105.11 - 2026-05-12
### Highlights
### Other Changes
## 0.105.10 - 2026-05-12
### Highlights
### Other Changes
## 0.105.9 - 2026-05-12
### Highlights
### Other Changes
## 0.105.8 - 2026-05-12
### Highlights
### Other Changes
#### Other
- Update AGENTs guide (c5b71a50)
#### Refactors
- Update installer scripts to fetch recent releases and improve error handling (1f45415f) (@vinhnx)
- Convert async functions to inline-delegating wrappers to reduce coroutine state machine overhead (5ea90914)
## 0.105.7 - 2026-05-07
### Highlights
### Other Changes
## 0.105.6 - 2026-05-03
### Highlights
### Other Changes
#### Other
- Refactor tool execution and configuration; add tool call delay tests and enhance agent capabilities (a8ce67f7)
## 0.105.5 - 2026-05-01
### Highlights
### Other Changes
#### Other
- Remove keyring dependency and add platform-specific keyring stores (fef8f0a2)
#### Refactors
- Reorganize imports and improve relaunch logic (91a841e1)
## 0.105.4 - 2026-05-01
### Highlights
### Other Changes
## 0.105.3 - 2026-05-01
### Highlights
### Other Changes
## 0.105.2 - 2026-04-30
### Highlights
#### Features
- Add tempfile dependency and enhance credential file permissions handling (6c198b3a)
- Enhance tool execution safety and prompt alignment handling (60c35814)
### Other Changes
#### Other
- Enhance OpenAI Response's API websocket provider logic and tests; clarify documentation and comments. (7368a946)
#### Refactors
- Replace direct file writes with write_private_file for secure permissions (0319c5e7)
## 0.105.1 - 2026-04-29
### Highlights
### Other Changes
#### Other
- Fix ast-grep dependencies install (21585180)
## 0.105.0 - 2026-04-27
### Highlights
### Other Changes
#### Other
- Add deepseek-v4-pro:cloud under Ollama provider (512f260b)
#### Refactors
- Streamline tool handling and validation logic, remove unused code (477319a0)
- Update skill manifest handling and permissions to use boxed types (aab2d60c)
## 0.104.2 - 2026-04-26
### Highlights
#### Features
- Update default model and add ast-grep tool support (d2599e7a)
- Enhance local agents functionality and improve demo scripts (fea4eaf2)
- Implement caching for lightweight skill metadata and system prompt layers (4ba51c73)
- Enhance ownership management and borrowing practices across tools and registries (c606e2d4)
- Add MCP lifecycle controls and related tool functionalities (c186f725)
- Update tool policies and available tools in tool-policy.json (cd016ae5)
### Other Changes
#### Other
- Add support for spawning background subprocesses and related configurations (17665666)
- Refactor configuration loading and improve error handling across modules (b7be8464)
## 0.104.1 - 2026-04-25
### Highlights
#### Bug Fixes
- Handle dated GPT-5.5 alias in model routing (34079b94)
### Other Changes
#### Other
- Add DeepWiki badge to README (7b21006a) (@vinhnx)
- Update config (59a5f22c)
## 0.104.0 - 2026-04-25
### Highlights
#### Features
- Update tool policies and default model configuration (fea8947a)
- Enhance tool call safety validation and refactor permission checks (87f47eec)
- Add tool budget warning handling and input reset functionality (bea41f6c)
- Add tool call validation with session limit handling in CopilotRuntimeHost (a4d420b6)
- Add support for GPT-5.5 model and update related configurations (3296c4ce)
### Other Changes
#### Refactors
- Streamline tool policies and update reasoning effort defaults for OpenAI models (15e635af)
- Reorganize AGENTS.md for clarity and improved workflow guidance (4732382c)
## 0.103.3 - 2026-04-24
### Highlights
#### Features
- Add Ollama cloud provider support for DeepSeek V4 Flash (92908fe3)
- Add DeepSeek V4 Pro and Flash cloud models support (973ec7ff)
### Other Changes
## 0.103.2 - 2026-04-24
### Highlights
### Other Changes
## 0.103.1 - 2026-04-24
### Highlights
### Other Changes
#### Other
- Update (e3590bd5)
## 0.103.0 - 2026-04-24
### Highlights
### Other Changes
#### Other
- Update (a4c12b76)
## 0.102.0 - 2026-04-24
### Highlights
### Other Changes
#### Other
- Update tests (896b1494)
## 0.101.0 - 2026-04-24
### Highlights
### Other Changes
#### Other
- DeepSeek V4 Models identifiers and descriptions to reflect version changes (be3638ba)
#### Refactors
- Change function signatures to return &str instead of &'static str (8b315ada)
## 0.100.3 - 2026-04-21
### Highlights
### Other Changes
## 0.100.2 - 2026-04-21
### Highlights
#### Features
- Add Kimi K2.6 support (d0a6a254)
- Add Kimi K2.6 to OpenRouter presets (04ca6808)
- Add MoonshotAI Kimi K2.6 model support (8882cf58)
### Other Changes
## 0.100.1 - 2026-04-21
### Highlights
#### Features
- Add Kimi K2.6 model support in Ollama and Hugging Face providers (dbb69fc0)
### Other Changes
## 0.100.0 - 2026-04-20
### Highlights
#### Bug Fixes
- Simplify metadata code retrieval in tests (5c4dbd5b)
#### Features
- Update Moonshot model references and configurations (365c522f)
- Update Moonshot model references to use DEFAULT_MODEL and add presets (794ee1a5)
### Other Changes
## 0.99.3 - 2026-04-20
### Highlights
#### Features
- Add request_user_input_enabled flag to system prompt context and update related logic (ad39062d)
### Other Changes
#### Refactors
- Convert async functions to return Future for improved performance (38fc4722)
- Replace manual string truncation with truncate_byte_budget utility for consistency (29a6dc34)
- Replace eprintln with tracing for consistent logging across modules (803235eb)
## 0.99.2 - 2026-04-19
### Highlights
#### Bug Fixes
- Upgrade rustls-webpki to 0.103.12 (#2 #3) (b53f9c9a) (@vinhnx)
- Update default model to claude-sonnet-4-6 across configurations and documentation (13f942bb)
#### Features
- Implement memories subsystem with configuration options (021aa664)
### Other Changes
#### Other
- Add OpenCode Zen and OpenCode Go providers and models (42bbea49)
- Refactor OpenCode providers to support model ID resolution and enhance compatibility (0a5bf2c1)
## 0.99.1 - 2026-04-17
### Highlights
#### Features
- Add handling for reasoning signature in UI and ACP streams (9049a09a)
### Other Changes
## 0.99.0 - 2026-04-16
### Highlights
### Other Changes
#### Other
- Update Anthropic provider to support Claude Opus 4.7 with new task budget and reasoning effort features (0a278b96)
## 0.98.7 - 2026-04-15
### Highlights
#### Bug Fixes
- Update condition for missing platform binaries in release script (3e88cd0e)
- Improve user prompt handling and update configuration settings (934ee662)
#### Features
- Add checksum generation and validation to release process (6708fec0)
- Add generic action required status handling in overlays (6d7ffdad)
### Other Changes
#### Other
- Update TODO (92336dc0)
- Update tool execution status display and refine UI handling (f77c7053)
- Implement pending user prompt handling and enhance status rendering in TUI (2c4dae7b)
## 0.98.6 - 2026-04-13
### Highlights
#### Features
- Add meta handling to elicitation response in LoggingClientHandler (0326ef68)
### Other Changes
## 0.98.5 - 2026-04-13
### Highlights
### Other Changes
#### Other
- Update deps (bd9e90ef)
## 0.98.4 - 2026-04-13
### Highlights
### Other Changes
## 0.98.3 - 2026-04-13
### Highlights
### Other Changes
## 0.98.2 - 2026-04-12
### Highlights
#### Bug Fixes
- Refine asset selection logic to prioritize .tar.gz over .sha256 files (f113d8a6) (@vinhnx)
#### Documentation
- Update AGENTS.md and TODO.md for clarity on development workflow (bd4f0647)
- Remove broken copy/yank text issue description from TODO.md (29ad997b)
### Other Changes
#### Other
- Add README files for new modules and update existing ones (5c4384be)
- Add subagent constants, discovery, model resolution, prompt handling, and types (63230b30)
- Add tests for transcript rendering, Vim mode functionality, and wizard interactions (1505cbfa)
- Update code structure for improved readability and maintainability (39917c56)
- Add persistent memory tests for memory operations and classification (a0e0a343)
- Refactor code structure for improved readability and maintainability (3ffdf8a4)
- Refactor code for improved readability and consistency; update default model in configuration (d42726a2)
- Update theme in configuration to 'ciapre-blue' (c12bdf36)
- Add tool outcomes handling for response content and subagent memory (1327986f)
- Add session archive tests for persistence and message handling (311bd727)
- Refactor tool names to use constants from the tools module for consistency (44181fa5)
- Add runtime tests for Copilot functionality in llm_request module (d038b0cf)
- Add comprehensive tests for tool execution and context handling (0d3b4776)
- Refactor code for improved readability and consistency in copilot_runtime, components, and skills modules (b2cc57e0)
- Add tests for structural search functionality and request validation (f4649c89)
- Add tests for tool execution and sandbox runtime behavior (cbf7effb)
- Refactor UI command item creation for improved readability and maintainability (0db3b4d4)
- Update tool policies, fix keyboard shortcuts, and enhance documentation (8b91a421)
- Add websocket active response error handling and tests (a784ca34)
- Refactor tool policies and enhance lifecycle hook payloads to include tool call ID (f27a37cd)
- Add support for new session trigger and refactor session initialization logic (7fe243b0)
- Add grouping functions for MCP tools by provider and update related logic (b58a72f9)
- Add warning handling for session archive persistence and finalization (7341d25e)
- Update descriptions for agent parameters and built-in tool registrations (69e86421)
- Update command preview notice to include sharing instructions (4588b09c)
- Refactor diagnostics and UI interaction to enhance instruction context handling (f1bcb2d4)
- Refactor JSON schema handling and enhance tool specifications (c719871f)
- Add meta field to MCP elicitation request and response structures (914d59b8)
- Add notification condition for desktop notifications delivery (559bba94)
- Add default target for docs.rs in Cargo.toml files (12a5d068)
- Refactor code for improved readability and consistency in multiple files (7cc15514)
- Remove outdated TODO items from project documentation (58aec7ee)
#### Refactors
- Enhance scrolling behavior and selection adjustment in session management (34060deb)
#### Tests
- Enhance nested directories test to filter file matches (93d8e397)
## 0.98.1 - 2026-04-09
### Highlights
#### Bug Fixes
- Prevent blocked handoff on tool-free recovery tool calls (f3d5a7ee)
- Make external editor opens transient and suppress escape leaks (1e8bd327)
- Update theme configuration to use homebrew style (4205a879)
- Update agent configuration and tool policies for improved functionality (f9cdc03d)
#### Features
- Enhance tool policies and workspace header signals for improved user experience (07910f3c)
### Other Changes
#### Other
- Refactor slash command messages and update guidance for installation (09028a6a)
- Implement status line setup and configuration, deduplicate terminal title items, and update task tracker permission in config (c4b7cb48)
- Implement fallback handling for failed tool-free recovery and adjust turn completion logic (4e58ac64)
- Implement structured timeout error handling and add tests for timeout scenarios (c214f667)
- Update tool policies, enhance exec command preparation, and improve documentation clarity (43c048e4)
- Commit (8ed8fb87)
- Bump version to 0.50.12 and remove chat-related commands and references (8a8efa6c)
- Bump version to 0.50.13 (337fe1ab)
- Implement subagent executors and enhance tool intent handling (c550faa2)
- Refactor exec session handling and add error reporting and tool search functionality (dc9e150f)
- Refactor tool handling and enhance session management for improved output processing (8d8c47ab)
- Refactor MCP provider initialization and enhance tool fetching logic (ecff001a)
- Refactor compaction thresholds and remove context awareness support from LLM providers (31b771d0)
- Update TODO (203ec0f3)
- Update tool policies, enhance recovery context previews, and improve structural search handling (a5db2b86)
- Enhance tool output handling and summary generation (c10d90fd)
- Refactor CLI action resolution and improve header display logic (61f15e54)
- Enhance CLI argument parsing and error reporting for workspace validation (77c3efbe)
#### Refactors
- Clean up tool policies and improve CLI help output (b32129ff)
## 0.98.0 - 2026-04-07
### Highlights
#### Features
- Add mouse pointer shape handling and link/text selection detection (3f654089)
- Implement clipboard copy functionality and add copy notification (5a28a4be)
- Strip underlines from local file links in inline UI and add related tests (728eadc6)
- Enhance CLI progress reporting with terminal clear functionality (a600a3b4)
- Add Z.AI GLM-5.1 model support across various providers and update related tests (645df2e9)
### Other Changes
#### Other
- Core split (3d12f6b6) (@poelzi)
- Update notes (f3ebfd31)
## 0.97.4 - 2026-04-07
### Highlights
### Other Changes
#### Other
- Fix formatting in README.md for clarity (979d991d) (@vinhnx)
- Update note (9f327f92)
#### Refactors
- Unify tool execution kernel and align harness safety limits (5fbfc7a8)
## 0.97.3 - 2026-04-07
### Highlights
### Other Changes
#### Other
- Fix tool batching and runloop regressions (2593f015)
- Consolidate tool fallback and rate-limit parsing (02fdb3c9)
#### Refactors
- Remove Vim mode command and related functionality (f3b44164)
- Simplify vtcode configuration (ab835726)
- Remove obsolete code (a55f265c)
- Remove contributors section from changelog template (30707f13)
- Remove FollowUp variant from ReplyKind and update related logic (6f12c2f7)
- Improve duration handling and validation in rate limiting and retry logic (3ec14baf)
## 0.97.2 - 2026-04-06
### Highlights
#### Bug Fixes
- Clear publish warnings (e5d2cb78)
### Other Changes
### Contributors
@vinhnx
## 0.97.1 - 2026-04-06
### Highlights
#### Features
- Enable ATIF trajectory export and update related configurations (47a230cc)
### Other Changes
#### Other
- Update TODO.md (824acefb) (@vinhnx)
### Contributors
@vinhnx
## 0.97.0 - 2026-04-05
### Highlights
#### Bug Fixes
- Include terminal-detection before tui publish (6f68a365) (@vinhnx)
- Publish specs crates before core (0c1da64d) (@vinhnx)
- Allow publishing specs crates (2f91bd14) (@vinhnx)
- Update task_tracker permission and model version in configuration (05405f5c) (@vinhnx)
- Harden perf baseline startup measurement (32f657dc)
- Animate active file operation indicator (c2baca0a)
- Enhance command safety checks and segment handling (d7491771)
- Dim top-level task tail lines in tool blocks (060ec1f5)
- Isolate shared state locks (0d685107)
#### Documentation
- Add Codex and debug startup release notes (9d24fb99)
#### Features
- Add Anthropic native memory and tool result clearing (7b7124f2)
- Update notification delivery mode to desktop (80b49f12)
- Combine native clearing with compaction (4edb1f65)
- Implement orchestrated harness with contract evaluation (b437d588)
- Enhance scorecard validation and add tests for missing/invalid criteria (29e3df0a)
- Update tool policies and add new tools with appropriate permissions (b7e51521)
### Other Changes
#### Other
- Fix badge duplication in README.md (4ffc53bb) (@vinhnx)
- Migrate to serde-saphyr (c29cdd5d) (@Sachin-Bhat)
- Enhance first-run wizard with startup modes and memory (199d1569)
- Fix check script and logging guard regressions (c6781996)
- Refactor system prompts and contracts for clarity and conciseness; optimize skill rendering and memory handling; enhance path policy checks; update execution kernel for unified file actions; adjust configuration formatting. (1b1d0c79)
- Remove sparse-checkout configuration from release workflow steps (d4be7f53)
- Enhance recovery handling and fallback messages; update command sharing options; adjust configuration for model and tool limits. (ebad1810)
- Update agent prompt for auto permission review rule reviewer and adjust model settings in configuration (7ae71a83)
- Add permission request and stop hooks to lifecycle management (5186a3ea)
- Refactor permission handling and improve code readability across multiple modules (a28db433)
- Implement notification system with /notify command and enhance session start hooks (e1bc7e41)
- Add configurable desktop notification backends (bf8c51d8)
- Enhance pty command handling, update configuration options, and add new JSON schemas for collaboration and command execution (6cf2dd04)
- Refactor vtcode.toml configuration: streamline agent settings, enhance permissions, and update context management (a5e4e440)
- Tighten persistent memory extraction and tool policies (f15b43a5)
- Add Agent Trajectory Interchange Format (ATIF) support for session trajectory export (0e09ef90)
### Contributors
@vinhnx
## 0.96.15 - 2026-04-04
### Highlights
### Other Changes
#### Other
- Update TODO (35cc8ccc)
### Contributors
@vinhnx
## 0.96.13 - 2026-04-04
### Highlights
### Other Changes
#### Other
- Update TODO (35cc8ccc)
### Contributors
@vinhnx
## 0.96.12 - 2026-04-02
### Highlights
#### Bug Fixes
- Encode SHA256 digest as hex manually (db7768f8)
- Encode config fingerprint digest as lowercase hex (4d470c5e)
### Other Changes
#### Other
- Fix script (5655308f)
- Refactor mock server initialization and error handling in tests; streamline MCP parameter building; update checksum calculation; enhance session rendering in TUI tests; adjust file permissions in configuration. (03d8ccc9)
### Contributors
@vinhnx
## 0.96.11 - 2026-04-02
### Highlights
#### Bug Fixes
- Encode SHA256 digest as hex manually (db7768f8)
### Other Changes
### Contributors
@vinhnx
## 0.96.10 - 2026-04-02
### Highlights
### Other Changes
#### Other
- Fix ast-grep init (0e622cd5)
- Fix TUI OAuth cancel without exiting VT Code (ce034434)
- Enhance modal rendering with title chrome and divider styles (339be092)
- Improve modal list dividers and highlight styling (803a63b5)
- Enhance modal rendering with improved styles and background handling (0b4278fb)
- Refactor palette rendering to include dividers and improve header styling (a4dde38f)
- Add Ghostty VT integration and update dependencies (53431d41)
- Add clear background rendering to floating modal and improve tests (c6b1afa0)
- Implement reborrow method in SlashCommandContext, update agents command usage, and change OpenAI provider settings in configuration. (2fb54da2)
- Update example blocked patterns in vtcode.toml.example (15c69b16)
- Enhance /compact command functionality and update configuration settings (8c31b153)
- Add freeform default handling to wizard steps and update related prompts (39926d0b)
### Contributors
@vinhnx
## 0.96.9 - 2026-03-31
### Highlights
#### Bug Fixes
- Make wizard modal `Enter` apply explicit freeform defaults and advance consistently with mouse submission
### Other Changes
### Contributors
@vinhnx
## 0.96.8 - 2026-03-31
### Highlights
### Other Changes
#### Other
- Fix release (ad544b2c)
### Contributors
@vinhnx
## 0.96.7 - 2026-03-31
### Highlights
### Other Changes
#### Other
- Add zread badge (6bce27c6)
- Refactor tests and update tool policies for improved functionality (2bf10861)
### Contributors
@vinhnx
## 0.96.6 - 2026-03-31
### Highlights
### Other Changes
### Contributors
@vinhnx
## 0.96.5 - 2026-03-31
### Highlights
#### Bug Fixes
- Correct post-release regressions (3cd9dde3)
### Other Changes
#### Other
- Update TODO (0b58348f)
### Contributors
@vinhnx
## 0.96.4 - 2026-03-31
### Highlights
#### Bug Fixes
- Pass model behavior to anthropic api provider (70944e75)
### Other Changes
### Contributors
@vinhnx
## 0.96.3 - 2026-03-31
### Highlights
### Other Changes
#### Other
- Update config (ce0e36d7)
### Contributors
@vinhnx
## 0.96.2 - 2026-03-31
### Highlights
#### Bug Fixes
- Unify external URL guardrails across VT Code flows (a6444c4b)
- Suppress loading UI for spool chunk reads (1635a52f)
#### Documentation
- Update README and scripts for crate publishing process (92c4275d)
#### Features
- Implement link click throttling and improve mouse event handling (f6746ce3)
- Implement ast-grep command checks and improve initialization process (11e1d04f)
### Other Changes
#### Other
- Update TODO.md (cb7ca5c1) (@vinhnx)
- Transform Rust source files (0e4c43d8) (@codemod[bot])
- Implement OAuth support for HTTP MCP providers and enhance CLI commands for login/logout (f6cd0c61)
- Add support for Codex app server integration and validation (a2a44dba)
- Update config (2d7bb992)
- Update TODO (76795760)
### Contributors
@vinhnx
## 0.96.1 - 2026-03-31
### Highlights
#### Bug Fixes
- Unify external URL guardrails across VT Code flows (a6444c4b)
- Suppress loading UI for spool chunk reads (1635a52f)
#### Features
- Implement link click throttling and improve mouse event handling (f6746ce3)
- Implement ast-grep command checks and improve initialization process (11e1d04f)
### Other Changes
#### Other
- Update TODO.md (cb7ca5c1) (@vinhnx)
- Transform Rust source files (0e4c43d8) (@codemod[bot])
- Implement OAuth support for HTTP MCP providers and enhance CLI commands for login/logout (f6cd0c61)
- Add support for Codex app server integration and validation (a2a44dba)
- Update config (2d7bb992)
### Contributors
@vinhnx
## 0.96.0 - 2026-03-30
### Highlights
#### Bug Fixes
- Update external editor instructions to reflect single-click behavior (7f9f0557)
#### Features
- Add shared lightweight model routing (4e41e088)
- Implement preferred lightweight model variants and update configuration settings (2112d993)
- Implement collect_single_response function for streamlined LLM interactions and refactor related code (a863561c)
- Add interactive OpenAI manual compact controls (6660de5d)
- Implement transcript link detection and wrapping functionality (9cfd3e01)
### Other Changes
#### Other
- Add parallel batch test for file listing and unified search; refactor list_files to avoid reentrancy (b5b2d2de)
- Add minimum turn limit for child agents and update documentation (97eb9a89)
- Update theme permissions and synchronize theme selection across components (43361475)
- Add theme management and accessibility features (512ad686)
- Add documentation for theme-related functions and improve code readability (9c387721)
- Enhance terminal restoration on panic and signal handling (d546487f)
- Implement scheduled tasks and automation framework (a0c65096)
- Add ast-grep integration for code quality checks and rule enforcement (a7b81d0a)
- Implement structural search workflows and enhance ast-grep integration (5112cdbb)
- Enhance ast-grep documentation and add tests for skill discovery queries (652ab8aa)
- Enhance ast-grep documentation and improve pattern syntax guidance (a5a71c9d)
- Enhance ast-grep documentation with rule cheat sheets and improve skill discovery tests (0e6a5a6c)
- Enhance ast-grep documentation with config cheat sheets and improve skill discovery tests (306f4ab4)
- Update TOOL_SPECS.md with strictness details and usage preferences; remove unused stats.rs; enhance SKILL.md and project-workflows.md with ast-grep insights; add tests for skill discovery and structural search flags; refine hints in structural_search.rs. (0cf298f6)
- Enhance documentation for custom language setup and tree-sitter integration; add tests for skill discovery by custom language queries. (71b6eadb)
- Enhance documentation on language injection and multi-language support; add tests for skill discovery by language queries. (16de0f58)
- Enhance ast-grep documentation and language support; add tests for new aliases and inference rules (644d908f)
- Enhance documentation for transformation objects and rewriters; add tests for skill discovery by new queries (211e21aa)
- Enhance documentation for sgconfig.yml and project workflows; add tests for skill discovery by ruleDirs, testConfigs, libraryPath, and dynamic injected queries. (4c0405d1)
- Enhance documentation and refine descriptions for ast-grep commands; normalize exit codes for no matches and findings in structural search. (29867cdc)
- Enhance documentation for Rust, TypeScript, TSX, YAML, Ruby, Python, Kotlin, Java, HTML, Go, Cpp, and C catalog examples; provide guidelines for adapting rules to repository-specific conventions. (819298a2)
- Enhance documentation for ast-grep project workflows and APIs; clarify project discovery behavior, scan vs run requirements, and performance optimizations (3db5c5e5)
- Fix typo in comments for line truncation functions (32ccac09)
- Add persistent memory support and enhance instruction appendix functionality (4c73f931)
- Refactor signal handling and improve session finalization timeout logic (47828a62)
- Refactor credential storage to support encrypted file storage and improve legacy migration handling (d2e97368)
- Add model configuration commands and lightweight model support (1458605d)
- Refactor transcript link handling and add URL guard feature (5463ab53)
### Contributors
@vinhnx
## 0.95.0 - 2026-03-30
### Highlights
#### Bug Fixes
- Update external editor instructions to reflect single-click behavior (7f9f0557)
#### Features
- Add shared lightweight model routing (4e41e088)
- Implement preferred lightweight model variants and update configuration settings (2112d993)
- Implement collect_single_response function for streamlined LLM interactions and refactor related code (a863561c)
- Add interactive OpenAI manual compact controls (6660de5d)
- Implement transcript link detection and wrapping functionality (9cfd3e01)
### Other Changes
#### Other
- Add parallel batch test for file listing and unified search; refactor list_files to avoid reentrancy (b5b2d2de)
- Add minimum turn limit for child agents and update documentation (97eb9a89)
- Update theme permissions and synchronize theme selection across components (43361475)
- Add theme management and accessibility features (512ad686)
- Add documentation for theme-related functions and improve code readability (9c387721)
- Enhance terminal restoration on panic and signal handling (d546487f)
- Implement scheduled tasks and automation framework (a0c65096)
- Add ast-grep integration for code quality checks and rule enforcement (a7b81d0a)
- Implement structural search workflows and enhance ast-grep integration (5112cdbb)
- Enhance ast-grep documentation and add tests for skill discovery queries (652ab8aa)
- Enhance ast-grep documentation and improve pattern syntax guidance (a5a71c9d)
- Enhance ast-grep documentation with rule cheat sheets and improve skill discovery tests (0e6a5a6c)
- Enhance ast-grep documentation with config cheat sheets and improve skill discovery tests (306f4ab4)
- Update TOOL_SPECS.md with strictness details and usage preferences; remove unused stats.rs; enhance SKILL.md and project-workflows.md with ast-grep insights; add tests for skill discovery and structural search flags; refine hints in structural_search.rs. (0cf298f6)
- Enhance documentation for custom language setup and tree-sitter integration; add tests for skill discovery by custom language queries. (71b6eadb)
- Enhance documentation on language injection and multi-language support; add tests for skill discovery by language queries. (16de0f58)
- Enhance ast-grep documentation and language support; add tests for new aliases and inference rules (644d908f)
- Enhance documentation for transformation objects and rewriters; add tests for skill discovery by new queries (211e21aa)
- Enhance documentation for sgconfig.yml and project workflows; add tests for skill discovery by ruleDirs, testConfigs, libraryPath, and dynamic injected queries. (4c0405d1)
- Enhance documentation and refine descriptions for ast-grep commands; normalize exit codes for no matches and findings in structural search. (29867cdc)
- Enhance documentation for Rust, TypeScript, TSX, YAML, Ruby, Python, Kotlin, Java, HTML, Go, Cpp, and C catalog examples; provide guidelines for adapting rules to repository-specific conventions. (819298a2)
- Enhance documentation for ast-grep project workflows and APIs; clarify project discovery behavior, scan vs run requirements, and performance optimizations (3db5c5e5)
- Fix typo in comments for line truncation functions (32ccac09)
- Add persistent memory support and enhance instruction appendix functionality (4c73f931)
- Refactor signal handling and improve session finalization timeout logic (47828a62)
- Refactor credential storage to support encrypted file storage and improve legacy migration handling (d2e97368)
- Add model configuration commands and lightweight model support (1458605d)
- Refactor transcript link handling and add URL guard feature (5463ab53)
### Contributors
@vinhnx
## 0.94.0 - 2026-03-28
### Highlights
#### Features
- Add hosted shell network policy support (def96551)
- Add support for OpenAI service tier "flex" and update related configurations (275f88c8)
- Add support for inline file inputs in OpenAI provider and enhance file handling (c07e0ac4)
### Other Changes
#### Other
- Refactor skills system: update documentation, remove version and author fields, streamline validation, and enhance template structure. (e6821250)
- Add vtcode-check tool and update related configurations and tests (ae563671)
- Remove outdated documentation and improve error messaging format (3f940788)
- Remove outdated documentation files and finalize release notes for v0.3.0 (2c2f01d8)
- Remove unused dependencies and improve session ID handling in executors (b89fc354)
- Refactor recovery handling and improve timeout messaging in execution policy (f0d3f512)
- Enhance Responses API support with continuation state management and incremental history handling (0f030292)
- Refactor auto permission review features and update related configurations (b9843702)
- Add diff preview functionality and related utilities (f98e7b7c)
- Refactor LLM request retry logic and add previous response chain error handling (0c18e268)
- Add support for minimax tool calls and enhance harmony tool parsing (d71bfeb7)
- Refactor minimax tool call parsing and add test for invocation without parameters (d50a6bae)
- Refactor LLM request retry logic and update provider configuration (b5e47b42)
- Refactor tool error handling and retry logic with structured ToolExecutionError (83e1fc97)
- Refactor tool error handling to use structured error context (5414b932)
- Refactor and consolidate code to reduce duplication across modules (deef3505)
- Add subagent controller management to ToolRegistry (c866f9b9)
- Add GLM-5.1 model with enhanced capabilities and update references (8b8882f7)
- Update .gitignore to include plans directory and tool-policy.json; add Codemod Skill Discovery section to AGENTS.md (67ba167f)
- Update TODO (1350cfe5)
- Add support for agent references, local agents management, and enhance sidebar rendering (cd51b91d)
- Refactor local agents handling and improve UI integration (f3bd979f)
#### Refactors
- Unify and streamline agent harness execution (cfcf3b05)
### Contributors
@vinhnx
## 0.93.3 - 2026-03-25
### Highlights
#### Bug Fixes
- Preserve compaction context in prompts and summaries (b38f1877)
#### Documentation
- Docs: update installation instructions for Homebrew and add troubleshooting note (0cb1c08d)
- Update architecture and contributing guidelines; add extension boundaries document (d17e267c)
#### Features
- Improve tool output handling by ensuring summary headers are always recorded and refining stdout removal logic for specific tool types. (83e59742)
### Other Changes
#### Other
- Update config (02384ca5)
- Add support for summarized session forks and update related logic (1ddfd7fd)
- Add inline prompt suggestion feature to chat composer (b813647f)
- Add agent loop contract documentation for VT Code lifecycle and events (f85428b6)
- Remove LiteLLM provider and related documentation (ca8889aa)
- Update TODO (f7b10332)
- Harden agent runner tool error handling (4671b804)
- Refactor error handling, optimize token usage, and enhance performance metrics (48af5696)
- Add streaming support OpenAI models with normalized event handling (ff76d3af)
- Add orchestration for plan-build-evaluate workflow with harness artifacts (6968408b)
#### Refactors
- Remove outdated prompt suggestion details and add subagent build task (c869b499)
- Enhance error handling and retry logic in tool execution (89075d3d)
- Improve error handling and classification in tool execution (4990c7dd)
- Enhance error handling and retry logic in agent execution (9f626e1e)
- Replace try_recv with recv for steering message handling (17edd05a)
- Introduce idle wake delay in inline event loop and related components (4fdec1dc)
- Streamline tool policy checks and improve error handling (25c11664)
- Replace RwLock with parking_lot::RwLock for pattern detector (278238fd)
- Add compact aliases for session_id and path parameters across tools (82340a2e)
- Enhance WebSocket handling for OpenAI provider, improve error detection, and update configuration documentation (9ae3e202)
- Enhance WebSocket handling and documentation for OpenAI provider (8df581c1)
- Streamline code structure and improve readability across multiple files (22d1fa22)
### Contributors
@vinhnx
## 0.93.2 - 2026-03-24
### Highlights
#### Bug Fixes
- Tighten prompt cache continuity and retention handling (7afbe9f0)
- Update sha256 checksum for macOS ARM architecture (39e5ace9)
- Update macOS ARM checksum and enhance Homebrew publishing script (58d1989a)
- Stabilize Zed ACP startup and transcripts (72f91134)
#### Documentation
- Add Rust style guidelines and update Clippy disallowed methods (5bedfa98)
- Clarify Zed ACP local development setup (266a146c)
#### Features
- Add configurable permissions policy (358f53fe)
- Add Claude-style compatibility for allowed and disallowed tools in permissions (bad5ba15)
### Other Changes
#### Other
- Update system prompt to specify use of plain ASCII icons for markers and bullets (3a8993f0)
- Add terminal management features and enhance session handling (8b07cb90)
- Update screen_backend.rs (9271147d)
#### Refactors
- Simplify key handling in modal state (bfec2be3)
- Reuse shared OpenAI prompt cache key helper (68fe597f)
- Tighten system prompts and document accuracy tuning (450dcef2)
- Update tool policies and enhance prompt suggestions in TODO (36313870)
### Contributors
@vinhnx
## 0.93.1 - 2026-03-22
### Highlights
#### Features
- Add thread safety primitives including `SendWrapper` and enable optional concurrency for native plugins via a `thread_safe` flag. (86c0c612)
### Other Changes
#### Other
- Delete checksums.txt (0d1ffdfd) (@vinhnx)
- Delete skills-lock.json (610fc7b7) (@vinhnx)
- Implement JSON-RPC-over-stdio transport for subprocess agents and refactor authentication methods (090bfc3f)
- Enhance ToolCall struct with optional namespace and update related parsing logic (ff97daba)
- Refactor RejectConfig documentation and update related tests for clarity (541eac42)
- Enhance RejectConfig documentation and update permission rejection logic (d7eb68f0)
- Add pod management functionality with model catalog and state handling (c029e115)
- Implement normalized streaming for OpenRouter provider and update response handling (b01aeb2f)
- Add prompt template support and enhance slash command handling (f23ff5c7)
- Prioritize modal key handling over copy-to-clipboard in process_key function (ff0a684a)
- Add MiniMax-M2.7 model support and update related configurations (70a0e776)
- Add tool invocation lifecycle events and refactor event recording (a67bf917)
- Refactor agent session controller and runtime for improved event handling and state management (9adda2cd)
- Refactor tool invocation handling and event recording in AgentRunner (9da2d85b)
- Refactor tool permission handling and streamline runtime steering integration (f91b0d32)
- Add run_standard_stream_attempt function and refactor streaming logic (09a5751f)
- Enhance tool permission handling, improve error messaging, and refine path suggestion logic (46521121)
- Add file read family tracking and recovery handling in tool outcomes (f88c2a3a)
- Implement glob pattern support for file listing and enhance related functionalities (62b7b831)
- Implement session-state management and approval caching for shell commands (eda2c70d)
- Implement transient UI components and refactor overlay handling (c7622e2d)
- Implement prompt cache lineage management and enhance runtime context handling (9e26d8de)
- Update session ID and timestamps in current_blocked.md; modify GPU type matching logic and add tests in catalog.rs; change API key and model settings in vtcode.toml (b5a2f974)
- Enhance system prompt handling by adding history directives and runtime context checks; add tests for new functionality (c406e88b)
- Refactor tool policies and update configuration for Ollama provider (eaaa2b83)
- Refactor logging levels from debug to trace for improved verbosity in various components (b8d75afb)
- Add history system directives handling and improve session configuration (d7f42900)
- Implement centralized logging and tracing system; add flush functionality and guidelines (8903d53c)
- Implement interaction support in GeminiProvider with error handling and streaming capabilities (2e1f34d8)
- Refactor path handling in tracing and debug logging; simplify flush logic in trace writer (2faf35a7)
- Refactor metrics handling to use ToolCatalogCacheMetrics struct for improved clarity and maintainability (495b5476)
- Refactor interrupt handling to ensure overlays are closed and improve user experience (6847f30f)
- Update TODO (8c9c3d55)
#### Refactors
- Remove explicit unsafe Send and Sync implementations for NativePlugin. (5c96c9d3)
### Contributors
@vinhnx
## 0.93.0 - 2026-03-22
### Highlights
#### Features
- Add thread safety primitives including `SendWrapper` and enable optional concurrency for native plugins via a `thread_safe` flag. (86c0c612)
### Other Changes
#### Other
- Delete checksums.txt (0d1ffdfd) (@vinhnx)
- Delete skills-lock.json (610fc7b7) (@vinhnx)
- Implement JSON-RPC-over-stdio transport for subprocess agents and refactor authentication methods (090bfc3f)
- Enhance ToolCall struct with optional namespace and update related parsing logic (ff97daba)
- Refactor RejectConfig documentation and update related tests for clarity (541eac42)
- Enhance RejectConfig documentation and update permission rejection logic (d7eb68f0)
- Add pod management functionality with model catalog and state handling (c029e115)
- Implement normalized streaming for OpenRouter provider and update response handling (b01aeb2f)
- Add prompt template support and enhance slash command handling (f23ff5c7)
- Prioritize modal key handling over copy-to-clipboard in process_key function (ff0a684a)
- Add MiniMax-M2.7 model support and update related configurations (70a0e776)
- Add tool invocation lifecycle events and refactor event recording (a67bf917)
- Refactor agent session controller and runtime for improved event handling and state management (9adda2cd)
- Refactor tool invocation handling and event recording in AgentRunner (9da2d85b)
- Refactor tool permission handling and streamline runtime steering integration (f91b0d32)
- Add run_standard_stream_attempt function and refactor streaming logic (09a5751f)
- Enhance tool permission handling, improve error messaging, and refine path suggestion logic (46521121)
- Add file read family tracking and recovery handling in tool outcomes (f88c2a3a)
- Implement glob pattern support for file listing and enhance related functionalities (62b7b831)
- Implement session-state management and approval caching for shell commands (eda2c70d)
- Implement transient UI components and refactor overlay handling (c7622e2d)
- Implement prompt cache lineage management and enhance runtime context handling (9e26d8de)
- Update session ID and timestamps in current_blocked.md; modify GPU type matching logic and add tests in catalog.rs; change API key and model settings in vtcode.toml (b5a2f974)
- Enhance system prompt handling by adding history directives and runtime context checks; add tests for new functionality (c406e88b)
- Refactor tool policies and update configuration for Ollama provider (eaaa2b83)
- Refactor logging levels from debug to trace for improved verbosity in various components (b8d75afb)
- Add history system directives handling and improve session configuration (d7f42900)
- Implement centralized logging and tracing system; add flush functionality and guidelines (8903d53c)
- Implement interaction support in GeminiProvider with error handling and streaming capabilities (2e1f34d8)
- Refactor path handling in tracing and debug logging; simplify flush logic in trace writer (2faf35a7)
- Refactor metrics handling to use ToolCatalogCacheMetrics struct for improved clarity and maintainability (495b5476)
- Refactor interrupt handling to ensure overlays are closed and improve user experience (6847f30f)
- Update TODO (8c9c3d55)
#### Refactors
- Remove explicit unsafe Send and Sync implementations for NativePlugin. (5c96c9d3)
### Contributors
@vinhnx
## 0.92.1 - 2026-03-21
### Highlights
#### Bug Fixes
- Remove idle input notification (367e91ba)
### Other Changes
#### Other
- Implement Ghostty VT backend for terminal emulation and snapshot rendering (2af5a11d)
- Refactor code for improved readability and performance; add loading indicators for better user experience (0d20012f)
- Add support for custom OpenAI-compatible providers in configuration (8a76debf)
### Contributors
@vinhnx
## 0.92.0 - 2026-03-21
### Highlights
#### Bug Fixes
- Remove repeated unrelated worktree prompts (8f93b519)
- Remove repeated unrelated worktree prompts (1f4ea2bd)
- Handle already cleared oauth logout states (081e20a6)
#### Documentation
- Update README and authentication guide to include GitHub Copilot support (2144d1d6)
### Other Changes
#### Other
- Update README.md (fa33fa70) (@vinhnx)
- Fix formatting in README for installation section (53fccf14) (@vinhnx)
- Update AgentSkills to Skills in README.md (00a37fa0) (@vinhnx)
- Update README.md (3aa6dcc2) (@vinhnx)
- Implement GitHub Copilot authentication and command handling (40da600d)
- Add request permission rejection handling and enhance Copilot integration (d713178d)
- Fix reasoning formatting in CopilotProvider to ensure proper newline handling (565ae957)
- Update notes (a66b4d84)
- Implement StdioTransport for JSON-RPC communication and refactor CopilotAcpClient to use it (543dbca1)
- Update note (5e7d9492)
- Refactor configuration initialization and improve code readability across multiple files; update API key and model settings in configuration. (dd5221bc)
- Update code structure for improved readability and maintainability (952b06fe)
- Add turn timeout management and recovery logic (bb180e82)
- Update TURN timeout recovery to use configured max execution time (e8f27aae)
- Refactor tool permission handling and add persistent approval cache (ca53f066)
#### Refactors
- Replace custom timing with /usr/bin/time for performance measurement (5be0c3b4)
- Update tool policies and improve command submission handling (890eacf0)
- Improve GitHub Copilot OAuth flow and update configuration settings (86b8d667)
### Contributors
@vinhnx
## 0.91.1 - 2026-03-20
### Highlights
#### Bug Fixes
- Auto-copy highlighted TUI selections (5a33748a)
- Copy selected input on command-c (05a4472f)
### Other Changes
#### Other
- Update TODO.md (a96c1aca) (@vinhnx)
- Add unrelated worktree prompt handling and related functionality (dd59dd65)
- Refactor model option handling and validation logic for improved performance and clarity (ed7d837a)
- Add OAuth authentication guide and enhance README with OAuth details (0660f149)
- Refactor system prompts and skills rendering for improved clarity and efficiency (7ba88e31)
#### Refactors
- Improve git diff preview logic and update related enums (18a30a87)
- Update prompt assembly logic and enhance instruction handling (8b6f71f5)
- Remove unused PromptAssemblyMode and simplify related code (1f348cd4)
- Streamline system prompt instructions for clarity and conciseness (dcf8fa34)
- Enhance performance guidelines and simplify related documentation (f6259c32)
### Contributors
@vinhnx
## 0.91.0 - 2026-03-18
### Highlights
#### Bug Fixes
- Ensure all OpenAI sessions are cleared on logout and fallback to API key (3d6cb033)
#### Features
- Enhance OAuth flow with clickable URLs and improved user guidance (d5e9afa9)
- Implement transcript link click handling and update keyboard protocol flags (12e48c6d)
- Enhance mouse selection with double-click detection and clipboard functionality (25bae2b5)
- Implement explicit copy request for mouse selection and update Homebrew publishing process (21f2f1da)
- Update styling for shimmer effect and change UI theme to ayu (53ca8d5e)
- Add MiniMax-M2.7 model and update related configurations (3eeb3664)
### Other Changes
#### Other
- Update TODO (a2eeed96)
#### Refactors
- Simplify model and provider selection entries, remove unnecessary enumerations (baea18fb)
- Streamline OpenAI session clearing by consolidating storage methods (23a6c55b)
### Contributors
@vinhnx
## 0.90.0 - 2026-03-17
### Highlights
#### Features
- Improve OAuth URL display by printing it on a separate line and remove related TODO item. (89d95a8a)
- Highlight file paths in markdown output (b50a3d62)
### Other Changes
#### Other
- Refactor session layout and rendering logic; consolidate bottom panel handling and improve input height management (e6e93dbf)
- Update format (b8d5aee6)
- Fix url wrap line (301f62c7)
- Refactor session handling and options; streamline overlay requests and keyboard protocol settings (bc30728c)
- Refactor input handling; consolidate trigger checks and update input commands (8c46dc45)
- Refactor input trigger updates; consolidate trigger checks into a single method (9c7b8562)
- Refactor core TUI frame rendering helpers (a59920bf)
- Refactor panel layout functions for file palette and history picker; streamline inline area calculations (0a83dff1)
- Add note to explore GitHub Copilot OAuth for using Anthropic's Claude (73ed2711)
- Add GPT-5.4 Nano and Mini models to OpenAI configuration (982b5c37)
- Add OpenAI tool search configuration and implement deferred loading policy (3a56178a)
- Add model management documentation and scripts for LLM integration (861bef97)
- Add ToolCatalogSource to tool registration and update related files (9db62221)
### Contributors
@vinhnx
## 0.89.2 - 2026-03-17
### Highlights
#### Bug Fixes
- Improve planning workflow entry and lifecycle tracking (13638d96)
- Improve argument and stderr preview truncation for safety (3a09f7c1)
#### Features
- Collapse large JSON outputs in TUI messages and adjust planning workflow policies to allow. (80eed12d)
### Other Changes
#### Other
- Refactor plan block parsing for safe character boundaries and add truncation for tool arguments and stderr previews (4b5a4b12)
- Fix read tool config issue (73f5cb6f)
- Update TODO (f7f2382f)
- Implement dynamic signature generation for task_tracker_create and add tests for payload changes (788215be)
- Refactor path and symbol hint extraction in research context collection (539c1ac3)
- Update TODO (cc9ee77d)
- Fix .gitignore entry for vtcode.toml and reorder configuration fields in vtcode.toml (d5e78f45)
- Fix planning workflow loading (708b5e99)
- Add plan confirmation prompt after draft persistence and enhance error classification (ccd75a8f)
- Enhance modal list state management to retain last selection and improve overlay behavior (22f813ce)
- Refactor planning workflow context handling and update configuration values for consistency (cc191c1d)
- Refactor plan validation logic to prefer more complete candidates and enhance related tests (f566129c)
- Enhance plan extraction and confirmation flow; add handling for incomplete drafts and improve task tracker management (7a2608f7)
- Refactor task tracker signature handling to use duplicate enforcement guard (f93d52a7)
- Update TODo (8d522e16)
- Update TODO (ac4b4f84)
- Refactor planning workflow handling and improve session loop logic (86b9b95c)
- Update TODO (ffdee65d)
- Update circuit breaker configuration and tests for improved resiliency (aa92cd91)
### Contributors
@vinhnx
## 0.89.1 - 2026-03-16
### Highlights
#### Features
- Enhance OpenAI Responses API handling and update configuration (5d5d55af)
- Enhance error handling by extracting human-readable messages from various JSON formats (b9d46cc2)
### Other Changes
#### Other
- Enhance OpenAI session handling and update UI status display (caa74f23)
- Implement recovery modes for empty responses and adjust tool call handling (e09d51c4)
- Add clipboard support for selected text and update default model and theme (439d5cd7)
- Refactor tool output handling to support spooled output references and improve clipboard functionality (0339abbd)
#### Refactors
- Simplify error sanitization logic and preserve full error content (7a3afa95)
### Contributors
@vinhnx
## 0.89.0 - 2026-03-15
### Highlights
#### Bug Fixes
- Add missing environment variables for Rust compilation in Dockerfile (4a51a588)
- Update success message and fallback logic for OpenAI API key (3895d00d)
- Enhance OpenAI OAuth handling and improve error reporting (9dd80633)
#### Features
- Add OpenAI hosted shell support and validation checks (46031529)
- Enhance tool permission handling with new prompt options and decision logic (d8f448ba)
- Add OpenAI hosted shell configuration and validation for skills (c0383c1d)
- Implement context-generic provider wiring and enhance provider configuration (6c5d1910)
- Add prompt metadata to SnapshotMetadata and enhance hydration logic (79ca0b4b)
- Enhance OpenAI provider with max output tokens and previous response ID handling (f4e5237e)
- Enhance LLM request handling with non-streaming support and structured history integration (4180821a)
- Update default model to gpt-5.3-codex and enhance reasoning handling in OpenAI provider (6f8b0c13)
- Add manual input support for OpenAI OAuth flow and enhance session handling (ad2107c2)
### Other Changes
#### Other
- Implement auto-compaction for turn loop and enhance skill templates with routing guidance (31aef8a1)
- Refactor plan file template for clarity and structure (df3764eb)
- Add support for blocked handoff events and quiet success output in hooks (f978598b)
- Update AGENTS instructions (b584d382)
- Add fallback filenames to ProjectDocOptions and related functions; enhance clickable link handling in ANSI renderer; implement session history persistence settings; introduce new FileOpener and HistoryPersistence enums in config. (3e4f1ea1)
- Add request_permissions field to RejectConfig and update related logic (54430912)
- Add support for explicit transcript links in the TUI session (31add726)
- Enhance exit summary output with model and reasoning details; update configuration for new provider and model (228f5909)
- Implement CGP integration facade for ToolRegistry with pipeline support (68bd7f63)
- Add direct tool completion handling and refactor related logic (ab246cb8)
- Scan commits for potential bugs (c6b7ac99)
- Implement mouse selection and interaction for modal lists and input areas (e618c0da)
- Enhance observability and model validation for LLM requests and tool executions (fc5c883a)
- Update documentation and improve terminal setup handling (6b7b82ea)
- Implement Vim mode support with key handling and text operations (e77c61e5)
- Implement Vim-style prompt editing engine with core functionality and tests (a722f56b)
- Add interactive features for prompt suggestions and job management (89c79954)
- Update prompt suggestion docs (aa3aaaae)
- Update TODO (d23690fc)
- Add OpenAI ChatGPT authentication support and update provider configuration (860a1d65)
- Add new GPT-5 Codex models and update related functionality (0cc7a335)
- Add client request ID handling and improve responses API integration (fb3220ac)
- Scan recent commits for bugs (dbb30d99)
- Enhance LLM response processing and metrics logging; update OpenAI model handling (bc8dd6aa)
#### Refactors
- Streamline OAuth handling and improve prompt rendering logic (91a3c73d)
- Enhance instruction rendering and highlight extraction in project documentation (cf116dac)
- Improve compaction handling and documentation across multiple files (fe86c4c5)
- Use get_config_dir for configuration directory retrieval in DotManager (28c82ef6)
- Replace provider resolution logic with infer_provider_from_model in create_provider_for_model (a8c208e3)
- Improve command processing and rendering efficiency in drive_terminal (e33ed10d)
- Update provider client creation logic and add debug logging for provider usage (9c947bde)
- Enhance conversation handling and user guidance in agent interactions (f16808a8)
- Enhance tool completion replies with LLM-generated suggestions and fallback logic (1ad0b3de)
- Integrate InlineHeaderContext into event handling and model selection (75c6bbc0)
- Consolidate session archive handling and improve debug log directory management (71f0b686)
- Reorganize imports and improve code formatting in interactive features (20817b85)
- Clean up code formatting and improve readability in multiple files (dffcbdae)
- Remove unused content delta handling in OpenAIStreamTelemetry (44da3f56)
- Remove PR review status handling and optimize TUI tick rates (737a2c75)
### Contributors
@vinhnx
## 0.88.5 - 2026-03-13
### Highlights
#### Features
- Add context awareness support and related tests for Anthropic models (0ff5066e)
- Add context window size support to header context and related components (ba2faa13)
- Implement effective context window size handling for models (7ab580b4)
### Other Changes
### Contributors
@vinhnx
## 0.88.4 - 2026-03-13
### Highlights
### Other Changes
#### Other
- Add support for tool-free recovery mode and enhance wait feedback mechanisms (8c9f4ca1)
- Refactor exit summary display for single and multi-model sessions; improve Ctrl+C signal handling for immediate exit (45558686)
- Enhance chat request parsing to preserve assistant phase and streamline message handling (cc3a923b)
- Enhance performance by optimizing vector initializations and memory allocations across multiple files (239ac831)
- Add 'tty' parameter for PTY execution and enhance output handling in exec responses (049c4382)
- Enhance error handling and output guidance for command execution failures (f1cdd4f2)
- Enhance error handling and recovery guidance for tool execution failures (1f3445c0)
- Refactor output handling to replace read_file and grep_file with unified_file and unified_search (5e094ba7)
- Replace read_file and grep_file references with unified_file and unified_search across multiple files (d9794cf4)
- Refactor unified_exec argument validation and enhance preflight checks (3d5c56c1)
- Refactor loop detection and recovery policies for unified execution (bb764846)
- Update continuation policy to default to "all" and implement pause/resume functionality in exec execution (c7b98951)
- Add default harness log directory and update event log path documentation (3135c322)
- Refactor input handling: update keybindings and queue submission logic (598a2ad5)
- Implement session IDE toggle (28c6a858)
- Add ANSI escape sequences for scroll region and insert/delete operations (339b6053)
- Update TODO (fe850c3d)
- Enhance execution flow: add non-interactive settling logic and improve test coverage (f7047954)
- Update TODO (45a02148)
- Add support for tracking and reusing successful readonly tool signatures (e97063ad)
- Add tool_call_id to event items and enhance usage tracking (7ab2319c)
- Add tool_call_correlation_id function for fallback handling in output items (134b5691)
- Enhance command detection: support likely typos and improve tool call correlation handling (2b64d276)
- Refactor ANSI handling: centralize constants, improve documentation, and enhance utility functions (d268d5c9)
### Contributors
@vinhnx
## 0.88.3 - 2026-03-12
### Highlights
### Other Changes
#### Other
- Overhaul model picker UI (b0f33444)
- Enhance prompt caching and telemetry metrics; update configurations for improved cache handling (3fb970ac)
- Update TODO (3ee104de)
- Implement task metadata handling and continuation logic in task tracker (96b8362a)
- Add effective_exec_events_path function and related tests; enhance harness_exec output normalization (05d86d58)
- Update notes (0664f0b9)
- Refactor tool output handling to remove follow-up prompts and streamline continuation logic (bd4000aa)
- Update note (52cdad90)
- Refactor execution result handling to improve metadata management and streamline output processing (a9ada48e)
- Add Nemotron-3-Super model support in Ollama provider and constants (2602a7fc)
- Enhance loop recovery hints and streamline error handling in tool execution (fabb03ed)
- Refactor tool output rendering to improve markdown handling and add hidden lines notice (62f7ad22)
- Refactor tool output rendering to streamline follow-up hints and improve output handling (51e754ac)
- Add support for Nemotron-3-Super model and improve terminal file path handling (33f04db1)
- Add exact terms match function and update default model to Nemotron-3-Super (8df4f93e)
- Add tests for file path handling and improve location suffix resolution (71638904)
- Add reference link for control sequences in TODO.md (d951810e)
- Remove empty code change entries from the changes log (fb89f6c2)
### Contributors
@vinhnx
## 0.88.2 - 2026-03-12
### Highlights
### Other Changes
#### Other
- Fix install script (25a0e060) (@vinhnx)
- Update badge colors and links in README.md (99175a26) (@vinhnx)
- Update README.md (b433aefb) (@vinhnx)
- Update TODO (163e9716)
- Refactor API key handling and update model reasoning capabilities (db0bdf03)
- Refactor reasoning handling in OpenRouterProvider and improve model checks (f237aa93)
- Refactor configuration loading to use workspace paths and remove deprecated environment variable handling (352cc9e5)
### Contributors
@vinhnx
## 0.88.1 - 2026-03-12
### Highlights
#### Features
- Add NVIDIA Nemotron 3 Super (free) model support (c3c203a0)
- Add live editor context and transcript file links (c330e4b3)
### Other Changes
#### Other
- Add search tools status badge and enhance language detection (7150057d)
#### Tests
- Add unit test for markdown table header rendering (db088ed7)
### Contributors
@vinhnx
## 0.88.0 - 2026-03-11
### Highlights
#### Features
- Refactor pattern detection logic, improve concurrency with `parking_lot`, and optimize performance with `SmallVec` and pre-allocated buffers. (28d0d534)
- Optimize tool execution tracking, file path parsing, and grep result handling, while preventing memory leaks in stats. (c319aebe)
- Add Miri check to the `check.sh` script and update documentation for detecting undefined behavior. (15f3d5c1)
### Other Changes
#### Other
- Refactor pattern detection and command safety checks for improved efficiency and optimize tool name caching. (4abdbc6f)
- Review ZedAgent registry mutex (3ead1b8b)
- Apply async docs updates (3b75f5cf)
- Incorporate async docs guidance (800f1927)
- Enhance task listing filters (24b35cee)
- Improve batched request handling (e1cbedf5)
- Improve ANSI styles (cbe58ce9)
- Refactor agent loop to support initial configuration and enhance live reload functionality (7aaba845)
- Review vtcode PR 13090 updates (e4ee2358)
- Refactor SandboxPolicy to support network access and allowlisting in ReadOnly mode (89127618)
- Enhance Anthropic tool support with code execution and memory tools; improve validation and error handling (fac76a15)
- Implement referenced Rust PR updates (c43428b1)
- Fix @file diff lookup (183a1ec3)
- Refactor tool routing and ACP (c37388a6)
- Add headings to vtcode configs (b0f294f7)
- Remove forced budget stop logic (6ec74d5c)
- Enhance interleaved thinking model support and serialization across providers (a8e8bd27)
- Refactor task tracker rendering and state management (06566014)
- Implement workspace environment variable management and interactive selection UI (7e94f5df)
- Add local tool execution and session management for ZedAgent (1a460fc2)
- Refactor session setup and tool execution: remove unused code and improve clarity (06cc7671)
- Implement ast-grep installer with caching and locking mechanisms; add ripgrep installer with platform-specific installation logic; enhance debug logging capabilities. (fd0dadc9)
- Clean unused rust code (98f12364)
- Refactor agent configuration: remove unused code, enhance API key environment handling, and streamline model selection logic (89cbd68a)
- Add Gemini wire models and streaming processor implementation (4114e15c)
- Refactor approval policy handling: streamline rejection logic and enhance clarity in related modules (10630d73)
- Refactor sandbox policy handling and improve test coverage (8bfd4efa)
- Refactor error handling and add fallback mechanisms for model not found scenarios in OpenAI and OpenResponses providers (12f7a717)
- Refactor provider configuration: add timeout and specific OpenAI/Anthropic options (288cb188)
- Implement file conflict monitoring and auditing features (07e4b493)
- Refactor code for improved readability and consistency across file operations and conflict detection (6354965c)
- Refactor file conflict handling and introduce runtime execution management (e60faf6a)
- Refactor update handling and relaunch logic; enhance update command execution strategy (dffb89da)
- Enhance tool policies and follow-up prompt detection; update theme configuration (1463e225)
- Update notes (aefceac2)
- Enhance TurnLoopContext and processing logic for improved tool call handling and metadata management (327379f2)
- Add exhausted_tool_call_limit method and update budget handling logic (ebdc1b6f)
- Add LiteLLM provider support and documentation (327183cb)
- Enhance parameter sanitization and alias handling for tool arguments (190ee131)
- Update TODO (3b2e19fc)
#### Refactors
- Introduce `detect_pattern_with_guard` for improved pattern detection logic and simplify `RwLock` write guard acquisition. (5f70ca7c)
### Contributors
@vinhnx
## 0.87.1 - 2026-03-08
### Highlights
### Other Changes
#### Other
- Enhance exec command functionality with session resumption and prompt handling (d234fee7)
- Add review command and related functionality (ef0298c7)
- Refactor tool execution and safety gateway: remove legacy orchestrator and unified executor, streamline parallel execution, and enhance safety context handling. (2946fcf7)
- Add support for resuming and forking sessions across workspaces (32bd498f)
- Add structured error handling with VtCodeError and ErrorCode (93beb517)
- Consolidate reliability plan (47186a92)
- Refactor ToolRegistration to use Arc<str> for name and add unregister_tool method in ToolRegistry (cba4392f)
- Refactor tool policy handling and enhance safety checks for file access (2e9a9c16)
- Refactor command session handling and rename PTY references to unified_exec (80023410)
- Refactor tool execution handling to unify command names and improve compatibility (e1f3b602)
- Refactor tool intent handling to unify command names and improve output processing (cb0581be)
- Refactor tool execution handling and update command aliases to use unified_exec (097b5d85)
- Add support for creating and managing PTY sessions in the tool registry (8cbfcb87)
- Add support for optional patch input in apply_patch handler and update related schemas (29ed7b6e)
- Add assistant phase handling to message processing and tests (f448df1c)
- Add support for OpenAI service tier configuration and handling (5586bffc)
- Implement VT Code ast-grep installer (f8901da1)
- Fix build release (f9f272e9)
#### Refactors
- Streamline event processing and enhance modal state management (1e7c6031)
### Contributors
@vinhnx
## 0.86.12 - 2026-03-07
### Highlights
#### Bug Fixes
- Preserve v prefix in GitHub release URLs for install script (3f2088be) (@vinhnx)
- Use exact tag names from GitHub API in install script (b22bd9af) (@vinhnx)
#### Features
- Add checksum for vtcode-0.86.11-aarch64-apple-darwin.tar.gz (47a2c486) (@vinhnx)
- Implement feature gating for tools and refactor related components (5f530f6d)
### Other Changes
#### Other
- Refactor session-state handling and update related components (d78397ef)
- Update TODO (ae3651f1)
- Update TODO (fb5ee65f)
- Refactor modal handling to use overlay system and improve session state management (d6171cfc)
### Contributors
@vinhnx
## 0.86.11 - 2026-03-06
### Highlights
#### Features
- Implement thread management and session handling improvements (e775bb58)
- Enhance session management with loaded skills and metadata handling (cb99499b)
### Other Changes
### Contributors
@vinhnx
## 0.86.10 - 2026-03-06
### Highlights
#### Features
- Implement thread management and session handling improvements (e775bb58)
- Enhance session management with loaded skills and metadata handling (cb99499b)
### Other Changes
### Contributors
@vinhnx
## 0.86.9 - 2026-03-06
### Highlights
### Other Changes
### Contributors
@vinhnx
## 0.86.8 - 2026-03-06
### Highlights
#### Features
- Enhance command execution tracking with arguments and output aggregation (80f4e11e)
- Refactor tool execution events to support new output and invocation structures (446a354c)
- Add tool output started event and integrate into streaming progress callback (d2551a50)
- Add support for OpenAI o3 and o4-mini models in Responses API (38f17b54)
- Enhance OpenAI Responses API support with hosted tools and validation (a17b42c8)
- Enhance shell execution approval process and protect writable metadata directories (74d40175)
### Other Changes
#### Other
- Update OpenAI model references to GPT-5.4 and adjust reasoning effort defaults (38c7644a)
- Implement shell command approval enhancements and justification handling (3ff40170)
#### Refactors
- Streamline ToolDefinition creation and improve serialization functions (5b3cc708)
### Contributors
@vinhnx
## 0.87.0 - 2026-03-06
### Highlights
#### Features
- Enhance command execution tracking with arguments and output aggregation (80f4e11e)
- Refactor tool execution events to support new output and invocation structures (446a354c)
- Add tool output started event and integrate into streaming progress callback (d2551a50)
- Add support for OpenAI o3 and o4-mini models in Responses API (38f17b54)
- Enhance OpenAI Responses API support with hosted tools and validation (a17b42c8)
- Enhance shell execution approval process and protect writable metadata directories (74d40175)
### Other Changes
#### Other
- Update OpenAI model references to GPT-5.4 and adjust reasoning effort defaults (38c7644a)
- Implement shell command approval enhancements and justification handling (3ff40170)
#### Refactors
- Streamline ToolDefinition creation and improve serialization functions (5b3cc708)
### Contributors
@vinhnx
## 0.86.7 - 2026-03-06
### Highlights
### Other Changes
### Contributors
@vinhnx
## 0.86.6 - 2026-03-05
### Highlights
#### Documentation
- Update documentation structure and add Development Setup guide (7b38475a)
### Other Changes
#### Other
- Refactor Cargo.toml files to use workspace dependencies and update versioning scheme (a7bf7eea)
- Add MCP requirements enforcement and update related configurations (1f39beec)
#### Refactors
- Modularize components in src/ for improved readability and maintainability (f405950c)
- Enhance plugin ABI with safety checks and optional deallocator (ef02885a)
- Integrate unicode-width for text display handling (f7251f5f)
- Implement schema command for runtime tool introspection and add input validation (8d2900a1)
- Improve visual distinction between assistant agent and PTY output messages (387a45c7)
- Update PTY fallback handling for consistent terminal output (ed0fa6c9)
### Contributors
@vinhnx
## 0.86.5 - 2026-03-04
### Highlights
#### Bug Fixes
- Update terminal notification handling for ghostty to use bell-only (ca512670)
#### Features
- Add Windows user profile root exclusions and enhance sensitive path matching (0a024067)
- Add new tools and policies for cargo and git commands (110d0326)
- Implement continuation arguments for tool execution and enhance read chunk handling (19db1b72)
### Other Changes
#### Other
- Update API key and model provider in vtcode.toml (2939fceb)
- Implement task guards for background operations and improve async handling in various modules (23a114ab)
- Implement interactive diagnostics with quick and full options in the slash command (6876ae63)
- Implement Markdown parsing with support for lists, links, and tables (0ded92d5)
- Implement prompt cache shaping mode and enhance caching strategies (5f43940d)
- Implement history normalization for LLM requests and enhance execution context management (ad977a32)
- Refactor tool output helpers and improve cache management (3c825982)
- Refactor configuration layer loading and error handling; add support for tracking layer origins in merges; implement API for reading and writing configuration with version checks. (5fe3d2da)
- Refactor shell command building, update justification messages, enhance search metrics, improve web fetch domain matching, and adjust tool policies in configuration. (3c5d7600)
- Refactor code for improved readability and consistency across multiple files (adb2ad30)
#### Refactors
- Streamline response handling and output printing in AgentRunner (d98a5e55)
- Simplify function output handling in HuggingFace and OpenAI providers (71381dd3)
- Enhance diff rendering styles and streamline background color handling (c85edbcf)
- Enhance color detection logic for terminal compatibility (af35cbf6)
- Update configuration section markers and adjust merging logic for user settings (79bce0e0)
- Enhance settings search functionality and improve nested key handling (81f799ca)
- Remove agent command and related logic, update slash command descriptions (d672e41a)
### Contributors
@vinhnx
## 0.86.4 - 2026-03-04
### Highlights
#### Bug Fixes
- Remove unused OpenRouter model variants and fix panic_hook warnings (d4afaa47)
- Update Cargo.toml to exclude all docs except CONFIG_FIELD_REFERENCE.md (6903748b)
### Other Changes
#### Other
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58)
- Implement interactive settings manager and refactor related components (1f2d7f2a)
- Update code structure for improved readability and maintainability (2ea69033)
- Update TODO (86150ea2)
- Update TODO (c0e49629)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf)
- Standardize error handling policies and improve retry logic across components (e4c248c9)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885)
- Implement interactive skills management and enhance planning workflow functionality (a0d19bc4)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7)
- Update TODO (45cc3a6a)
- Enhance interactive skills management with new commands and aliases (471cd578)
- Refactor slash commands and improve skills management interface (f0391983)
- Update TODO (c7c11e12)
- Update TODO (f03eb648)
- Enhance Planning workflow interview handling and UI adjustments (77b8c823)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509)
- Refactor error handling and improve code safety across multiple modules (74ee2dea)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f)
- Add native plugin support and enhance skill loading (8ba9fa9e)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b)
- Add native plugin system implementation and documentation (0c0c856b)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7)
- Add model_search_value function for generating model search strings (7ba202a8)
- Remove production.py and version.rs files from the project (f30fa2e2)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3)
- Refactor code structure for improved readability and maintainability (43dd252e)
- Refactor imports and update theme in configuration (a9517434)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68)
- Add memchr dependency and update usage in ANSI parser (1b16a714)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e)
- Refactor imports and update configuration settings across multiple files (05d95e7b)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b)
- Fix path (cc0eaa6a)
### Contributors
@vinhnx
## 0.86.3 - 2026-03-04
### Highlights
#### Bug Fixes
- Remove unused OpenRouter model variants and fix panic_hook warnings (d4afaa47)
### Other Changes
#### Other
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58)
- Implement interactive settings manager and refactor related components (1f2d7f2a)
- Update code structure for improved readability and maintainability (2ea69033)
- Update TODO (86150ea2)
- Update TODO (c0e49629)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf)
- Standardize error handling policies and improve retry logic across components (e4c248c9)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885)
- Implement interactive skills management and enhance planning workflow functionality (a0d19bc4)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7)
- Update TODO (45cc3a6a)
- Enhance interactive skills management with new commands and aliases (471cd578)
- Refactor slash commands and improve skills management interface (f0391983)
- Update TODO (c7c11e12)
- Update TODO (f03eb648)
- Enhance Planning workflow interview handling and UI adjustments (77b8c823)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509)
- Refactor error handling and improve code safety across multiple modules (74ee2dea)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f)
- Add native plugin support and enhance skill loading (8ba9fa9e)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b)
- Add native plugin system implementation and documentation (0c0c856b)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7)
- Add model_search_value function for generating model search strings (7ba202a8)
- Remove production.py and version.rs files from the project (f30fa2e2)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3)
- Refactor code structure for improved readability and maintainability (43dd252e)
- Refactor imports and update theme in configuration (a9517434)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68)
- Add memchr dependency and update usage in ANSI parser (1b16a714)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e)
- Refactor imports and update configuration settings across multiple files (05d95e7b)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b)
- Fix path (cc0eaa6a)
### Contributors
@vinhnx
## 0.86.2 - 2026-03-04
### Highlights
#### Bug Fixes
- Remove unused OpenRouter model variants and fix panic_hook warnings (d4afaa47)
### Other Changes
#### Other
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58)
- Implement interactive settings manager and refactor related components (1f2d7f2a)
- Update code structure for improved readability and maintainability (2ea69033)
- Update TODO (86150ea2)
- Update TODO (c0e49629)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf)
- Standardize error handling policies and improve retry logic across components (e4c248c9)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885)
- Implement interactive skills management and enhance planning workflow functionality (a0d19bc4)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7)
- Update TODO (45cc3a6a)
- Enhance interactive skills management with new commands and aliases (471cd578)
- Refactor slash commands and improve skills management interface (f0391983)
- Update TODO (c7c11e12)
- Update TODO (f03eb648)
- Enhance Planning workflow interview handling and UI adjustments (77b8c823)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509)
- Refactor error handling and improve code safety across multiple modules (74ee2dea)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f)
- Add native plugin support and enhance skill loading (8ba9fa9e)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b)
- Add native plugin system implementation and documentation (0c0c856b)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7)
- Add model_search_value function for generating model search strings (7ba202a8)
- Remove production.py and version.rs files from the project (f30fa2e2)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3)
- Refactor code structure for improved readability and maintainability (43dd252e)
- Refactor imports and update theme in configuration (a9517434)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68)
- Add memchr dependency and update usage in ANSI parser (1b16a714)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e)
- Refactor imports and update configuration settings across multiple files (05d95e7b)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b)
### Contributors
@vinhnx
## 0.86.1 - 2026-03-04
### Highlights
#### Bug Fixes
- Remove unused OpenRouter model variants and fix panic_hook warnings (d4afaa47)
### Other Changes
#### Other
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58)
- Implement interactive settings manager and refactor related components (1f2d7f2a)
- Update code structure for improved readability and maintainability (2ea69033)
- Update TODO (86150ea2)
- Update TODO (c0e49629)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf)
- Standardize error handling policies and improve retry logic across components (e4c248c9)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885)
- Implement interactive skills management and enhance planning workflow functionality (a0d19bc4)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7)
- Update TODO (45cc3a6a)
- Enhance interactive skills management with new commands and aliases (471cd578)
- Refactor slash commands and improve skills management interface (f0391983)
- Update TODO (c7c11e12)
- Update TODO (f03eb648)
- Enhance Planning workflow interview handling and UI adjustments (77b8c823)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509)
- Refactor error handling and improve code safety across multiple modules (74ee2dea)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f)
- Add native plugin support and enhance skill loading (8ba9fa9e)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b)
- Add native plugin system implementation and documentation (0c0c856b)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7)
- Add model_search_value function for generating model search strings (7ba202a8)
- Remove production.py and version.rs files from the project (f30fa2e2)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3)
- Refactor code structure for improved readability and maintainability (43dd252e)
- Refactor imports and update theme in configuration (a9517434)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68)
- Add memchr dependency and update usage in ANSI parser (1b16a714)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e)
- Refactor imports and update configuration settings across multiple files (05d95e7b)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b)
### Contributors
@vinhnx
## 0.86.0 - 2026-03-04
### Highlights
### Other Changes
#### Other
- Enhance reasoning detail handling across providers and improve model detection (ef1f5a59)
- Refactor model constants and remove deprecated models; update provider configurations and capabilities (3ba46998)
- Update configuration for agent provider and API key; enhance user settings guidance (b11c56e0)
- Refactor model constants for clarity and update provider logic for consistency (cc37511d)
- Enhance reasoning visibility handling and integrate detailed reasoning extraction across modules (d1d19a58)
- Implement interactive settings manager and refactor related components (1f2d7f2a)
- Update code structure for improved readability and maintainability (2ea69033)
- Update TODO (86150ea2)
- Update TODO (c0e49629)
- Refactor settings handling and enhance input tokenization for improved functionality (52e07688)
- Adjust loop detection parameters and enhance normalization for read-file tools (c651bfaf)
- Standardize error handling policies and improve retry logic across components (e4c248c9)
- Refactor recovery flow and error handling for improved diagnostics and recovery strategies (7a036885)
- Implement interactive skills management and enhance planning workflow functionality (a0d19bc4)
- Implement interactive skills manager in TUI similar to MCP manager (8a518681)
- Enhance plan confirmation flow with TUI integration and interactive options (affa91c7)
- Update TODO (45cc3a6a)
- Enhance interactive skills management with new commands and aliases (471cd578)
- Refactor slash commands and improve skills management interface (f0391983)
- Update TODO (c7c11e12)
- Update TODO (f03eb648)
- Enhance Planning workflow interview handling and UI adjustments (77b8c823)
- Update MiniMax provider to use Anthropic Messages API and adjust default model configuration (b21c5509)
- Refactor error handling and improve code safety across multiple modules (74ee2dea)
- Refactor TODO items for improved clarity and remove obsolete files (3ef5c370)
- Refactor tool intent classification and update tests for non-destructive tools (edb8b93f)
- Add native plugin support and enhance skill loading (8ba9fa9e)
- Refactor diff algorithms and improve bounds-checking in various modules (bcdaa19b)
- Add native plugin system implementation and documentation (0c0c856b)
- Add streaming progress events and enhance plugin handling in CLI (8ecb5670)
- Refactor TURN module and enhance UI interaction streaming (d20cb7dc)
- Update Hugging Face model constants and descriptions; normalize model ID handling for Step 3.5 Flash (261a6b0f)
- Enhance error formatting for unsupported models in HuggingFaceProvider (7703edd7)
- Add model_search_value function for generating model search strings (7ba202a8)
- Remove production.py and version.rs files from the project (f30fa2e2)
- Refactor recovery flow: streamline tab handling and improve question line processing (788d5300)
- Update default model in vtcode.toml to Step-3.5-Flash (915a20f3)
- Refactor code structure for improved readability and maintainability (43dd252e)
- Refactor imports and update theme in configuration (a9517434)
- Replace std::collections::HashMap with hashbrown::HashMap across the codebase for improved performance. (aa277f68)
- Add memchr dependency and update usage in ANSI parser (1b16a714)
- Add bytes dependency and update output handling in process and pipe modules (0fe2017e)
- Refactor imports and update configuration settings across multiple files (05d95e7b)
- Refactor caching implementation: migrate to UnifiedCache and remove deprecated LruCache (ba04ef9b)
### Contributors
@vinhnx
# Changelog
All notable changes to vtcode will be documented in this file.
## 0.85.3 - 2026-03-03
### Highlights
#### Features
- **Native Plugin System**: Implement native code plugin support using libloading for high-performance, pre-compiled skills
- Add `vtcode-core/src/skills/native_plugin.rs` with PluginLoader, NativePlugin, and NativePluginTrait
- Support for dynamic library loading (.dylib, .so, .dll) with platform-specific handling
- Plugin ABI with version checking and metadata validation
- Trusted directory model for secure plugin discovery
- Integration with EnhancedSkill enum for unified skill handling
- Comprehensive documentation and example plugin (hello-world)
- Add libloading 0.8 dependency to workspace and vtcode-core
### Documentation
- Add Native Plugin Guide (`docs/skills/NATIVE_PLUGIN_GUIDE.md`) - Complete user guide for plugin development
- Add Native Plugin Implementation Summary (`docs/skills/NATIVE_PLUGIN_IMPLEMENTATION.md`) - Technical implementation details
- Add example plugin with tests (`examples/plugins/hello-world/`)
### Other Changes
- Update SkillRoot structure with is_plugin_root field for plugin directory detection
- Add plugin discovery to skill loading system
- Add plugin validation and safety checks
- Update CHANGELOG with plugin feature
## 0.85.2 - 2026-03-02
### Highlights
#### Bug Fixes
- Use RngExt trait for random_range method in slug.rs (b9b03932)
- Trim whitespace from custom config and data directory environment variables (179f1ef9)
#### Documentation
- Add final notes and principles for coding agents in AGENTS.md (9d2e061c)
#### Features
- Include git version info in long_version and add tests (8a1fe108)
- Add snapshot files for history picker and slash widget tests (b397909d)
- Implement TUI log capture feature and enhance panic hook handling (89813122)
- Integrate color-eyre for enhanced error reporting in debug mode (05591670)
### Other Changes
#### Other
- Enhance skill authoring and routing logic; add response chaining support (121a8ece)
- Add `thiserror` integration for improved error handling across multiple modules (b5081d89)
- Add websocket connection limit error handling and update API versioning (6f76cfb7)
- Add installation scripts to release artifacts and update installation documentation (816c6d41)
- Refactor error log collector to use a bounded buffer and enhance link rendering in markdown (c7ef605b)
- Refactor line truncation logic and introduce utility module for consistent ellipsis handling (69742626)
- Refactor websocket header insertion and update apply_patch handler to use Arc for session and turn context (13e4b403)
- Add error log buffer size limit and refactor tool context for constrained policies (f8b71916)
- Refactor tool routing and permission handling; improve URL-aware text wrapping (564fdad7)
- Refactor tool permission caching and enhance error handling; improve text wrapping utilities (ed68b7fd)
- Add support for zsh exec wrapper and enhance PTY configuration (9d992fed)
- Add data-oriented design plan and optimization phases for vtcode (98175726)
- Enhance status line configuration and improve token extraction logic for file references (68bb39d4)
- Document applied default status line items pattern and outline data-oriented design plan for vtcode (db98b328)
- Add tests for PtyConfig validation and refactor session creation methods (8cd350ef)
- Refactor TODO.md to improve formatting and clarify implementation instructions (265c835a)
- Enhance file palette key handling to manage selection and normal submit behavior (309e8b22)
- Refactor process group termination logic for robust child process cleanup (bb1ab07f)
- Refactor WebSocket session handling and enhance incremental input validation (e874c269)
- Refactor sandbox permissions and enhance tool execution handling (f997edb2)
- Implement fine-grained rejection controls for approval prompts and update related logic (3b9a7775)
- Refactor caching mechanisms to use inner struct with RwLock for improved concurrency (46102de4)
- Refactor session initialization to encapsulate execution and metadata contexts (ef3d1020)
- Enhance TurnProcessingContext with new as_turn_processing_context method for improved context management (c0b47e4d)
- Refactor RunLoopContext and TurnProcessingContext to use constructor methods for improved clarity and maintainability (cd9d63a5)
- Refactor TurnProcessingContext to use parts for better modularity and readability (b133ff3d)
- Refactor code for improved readability and consistency in multiple files (2565bbdb)
- Add Clippy configuration for vtcode with custom lint settings (fac8a298)
- Implement theme synchronization for Alacritty, Ghostty, and Kitty terminals; remove CachedStyleParser and replace with LrMap for improved performance; add terminal color probe for enhanced theme detection; update Cargo.toml files for Rust 1.85 compatibility. (c7dd0f28)
- Refactor code to use `parking_lot::RwLock` instead of `std::sync::RwLock` for improved performance; update various imports and method calls across multiple files; bump Rust version to 1.88 in Cargo.toml files. (10892d08)
- Refactor code for improved readability and consistency across multiple files (d876e596)
- Refactor command processing, improve output handling, and enhance telemetry reporting (ec20b546)
- Refactor tests and improve code readability (2c2ade2b)
- Add color policy module and integrate into styling and error display (3287a044)
- Add fuzzing adoption plan for security parsers and initial setup (4b4d8d8f)
- Refactor code for consistency and readability across multiple modules (502cda47)
- Refactor middleware chain initialization to use push method (2368dec1)
- Update integration tests to replace 'cargo test' with 'cargo check' and adjust access count assertions (862c07df)
- Add fuzzing support for security parsers and update documentation (bd445376)
- Refactor command safety checks and improve loop detection tests (b8ede10b)
- Enhance command validation and tool execution with unified action handling (0c3d4b86)
#### Refactors
- Remove unnecessary Clear widget calls in header and input rendering (fb1c8de7)
- Streamline architecture documentation and enhance clarity (565ea315)
- Rename color-eyre hook functions for clarity and consistency (3c555a86)
- Update approval policy handling to use Arc<RwLock> for thread safety (3b91595f)
- Create sync_mcp_approval_policy_for_context function for improved clarity (92a6cf82)
- Improve error handling and simplify string conversion in terminal color probe and LM Studio client (e41acb42)
- Update exit summary formatting and remove unused functions (7c58ff16)
- Streamline lint configuration and enhance exit summary output (eaca32a6)
- Enhance exit summary formatting and improve file indexing logic (d143fe97)
### Contributors
@vinhnx
## 0.85.1 - 2026-03-01
### Highlights
### Other Changes
#### Other
- Enhance skill authoring and routing logic; add response chaining support (121a8ece)
- Add `thiserror` integration for improved error handling across multiple modules (b5081d89)
- Add websocket connection limit error handling and update API versioning (6f76cfb7)
- Add installation scripts to release artifacts and update installation documentation (816c6d41)
- Refactor error log collector to use a bounded buffer and enhance link rendering in markdown (c7ef605b)
- Refactor line truncation logic and introduce utility module for consistent ellipsis handling (69742626)
- Refactor websocket header insertion and update apply_patch handler to use Arc for session and turn context (13e4b403)
- Add error log buffer size limit and refactor tool context for constrained policies (f8b71916)
- Refactor tool routing and permission handling; improve URL-aware text wrapping (564fdad7)
- Refactor tool permission caching and enhance error handling; improve text wrapping utilities (ed68b7fd)
- Add support for zsh exec wrapper and enhance PTY configuration (9d992fed)
- Add data-oriented design plan and optimization phases for vtcode (98175726)
- Enhance status line configuration and improve token extraction logic for file references (68bb39d4)
- Document applied default status line items pattern and outline data-oriented design plan for vtcode (db98b328)
- Add tests for PtyConfig validation and refactor session creation methods (8cd350ef)
- Refactor TODO.md to improve formatting and clarify implementation instructions (265c835a)
- Enhance file palette key handling to manage selection and normal submit behavior (309e8b22)
- Refactor process group termination logic for robust child process cleanup (bb1ab07f)
- Refactor WebSocket session handling and enhance incremental input validation (e874c269)
- Refactor sandbox permissions and enhance tool execution handling (f997edb2)
- Implement fine-grained rejection controls for approval prompts and update related logic (3b9a7775)
- Refactor caching mechanisms to use inner struct with RwLock for improved concurrency (46102de4)
- Refactor session initialization to encapsulate execution and metadata contexts (ef3d1020)
- Enhance TurnProcessingContext with new as_turn_processing_context method for improved context management (c0b47e4d)
- Refactor RunLoopContext and TurnProcessingContext to use constructor methods for improved clarity and maintainability (cd9d63a5)
- Refactor TurnProcessingContext to use parts for better modularity and readability (b133ff3d)
- Refactor code for improved readability and consistency in multiple files (2565bbdb)
### Contributors
@vinhnx
## 0.85.0 - 2026-03-01
### Highlights
#### Bug Fixes
- Update contributor handling in changelog template (f1ef0a2c)
#### Features
- Add diagnostics visibility toggle in transcript for debug builds (2b339962)
### Other Changes
#### Other
- Refactor diff rendering and styling for consistency and accessibility (3edcf204)
- Refactor token counting methods and remove unused tests across providers; update configuration for default provider and model. (55c76984)
- Add health check and reconnection methods to MCP provider; enhance error classification and guidance (45fcc200)
- Enhance error handling and logging across providers; improve session archive retention and diagnostics (ce94ca83)
- Refactor error recovery to use tokio::sync::RwLock; update tool outcome handling and improve cache retrieval methods (76c1789f)
- Add support for "show diff" command and improve tool call validation (62051691)
- Update color scheme for unified diff background and gutter colors; change terminal theme to improve readability (525653e9)
- Add diff theme configuration and improve styling for terminal output (501634f1)
- Enhance interactive mode with new keyboard shortcuts and shell mode indicators; update configuration to disable turn timer (87919849)
- Refactor input handling and padding logic; enhance shell command validation and error feedback (d520d2ea)
- Add file reference metadata handling and update input with aliases (91e39103)
- Implement postamble exit summarization with metric calculation and enhance markdown output styling (02d01e99)
- Implement code change tracking and exit summary reporting (5e776910)
- Refactor session handling to streamline resume logic and enhance message rendering (6654f328)
#### Refactors
- Rename sessions command to resume and update related handlers (c63ab822)
- Enhance panic handling, optimize agent harness, and improve token efficiency (c68e7ea1)
- Enhance error handling and recovery mechanisms; improve error categorization and logging (43a4e099)
### Contributors
@vinhnx
## 0.84.1 - 2026-02-28
### Highlights
### Other Changes
#### Other
- Update code structure for improved readability and maintainability (24d81c35)
- Refactor themes and improve UI handling (970dadf8)
- Add support for native web search tools in Anthropic API (84c50dfb)
- Add support for file content parts in message handling and OpenAI provider (4effd39c)
- Update theme to "ciapre" across configurations and add new themes to the theme registry (9c8d0c16)
- Refactor syntax highlighting implementation and improve theme handling (a9236cfc)
- Enhance tool routing and UI theme accessibility (0747f3d8)
- Enhance diff rendering and suppress redundant recap in TUI (0f3513b1)
- Fix light theme rendering issues and improve accessibility; enhance theme handling and live reload functionality (371a681b)
- Normalize unified search arguments and enhance argument extraction logic (a732a449)
- Fix light theme rendering issues and improve theme handling for better accessibility (ec131237)
### Contributors
## 0.84.0 - 2026-02-28
### Highlights
#### Bug Fixes
- Strip x-extension fields from tool schemas and handle empty safety blocks (7ff0b66f)
#### Features
- Add terminal notifications for agent events and syntax highlighting for bash commands in TUI (d54ba169)
### Other Changes
#### Other
- Update badge links in README.md (b74377e3) (@vinhnx)
- Merge pull request #607 from LiGoldragon/mentci-gemini-json-fix (db195063) (@vinhnx)
- Refactor code for improved readability and performance; simplify conditionals and remove unnecessary clones (93989d58)
- Add notification system with configurable preferences and delivery modes (e3be6b6b)
- Remove deprecated tool aliases and clean up related code for user input handling (5a23ab29)
- Unify ModelId and Provider across vtcode-config and vtcode-core; add regression tests to prevent future drift (3bef6a4d)
- Remove deprecated model ID and provider configurations, streamline model handling and update references across the codebase. (ebd573ec)
- Improve (d8ec16e4)
- Refactor code for improved readability and maintainability; streamline model handling and enhance diff preview functionality. (58607c7b)
- Add Qwen3.5-397B-A17B model support and update related identifiers and descriptions (a9e62f9c)
- Add turn elapsed timer feature and configuration options (f6867385)
- Remove xlsx skill and related files; update turn outcome message format (c31af51e)
- Implement exact token counting for various LLM providers and enable token counting feature (136c1b1f)
- Remove unused question-related policies and enhance token usage logic in ContextManager (4e39398a)
- Remove xAI Grok models and related configurations from the codebase (7b069620)
- Enhance file reference extraction to handle npm scoped packages and improve path validation (00646152)
- Enhance UI styling for PTY output and improve color handling across components (36cc1ca0)
- Add sidebar, slash command palette, and transcript widgets; implement session options and host adapter traits (fdc70374)
- Add utilities for ANSI parsing, caching styles, and diff generation (a20742ed)
- Update configuration for agent provider and default model; adjust theme settings (465ed6e7)
- Enhance changelog generation with Highlights and Contributors sections (8e25479e)
- Enhance tool output handling and caching logic (8635f4a0)
- Refactor unified_exec tool to add inspect action and improve documentation (eac81276)
- Refactor guidance messages for clarity and conciseness; update tool output handling and error messages. (ae454f4b)
- Add 'continue' action to unified_exec tool; refactor related logic and documentation (b2102ff4)
- Refactor tool policies and enhance shell command parsing; add next_continue_args handling (327a2467)
- Add clear screen and copy latest assistant reply commands; update related documentation and tests (39b2d94a)
- Enhance command detection by normalizing natural language requests and adding support for various command patterns (bc2be0b9)
- Change terminal theme from 'ciapre-dark' to 'mono' (6b03347c)
- Add appearance configuration and improve modal handling (afd21755)
- Enhance external editor integration and rewind functionality (e5f722ce)
- Refactor code for improved readability and functionality; normalize unified search arguments; enhance validation error handling; update default model in configuration. (56adb18f)
- Refactor palette handling and LLM retry logic; improve error handling and tests (64f0f949)
- Enhance task tracker functionality and error handling (42781438)
- Remove XAI provider tests and related assertions from model validation (880ec280)
- Remove deprecated integration and workflow tests; clean up test module imports (3c9f5f42)
- Update model references from "gemini-3-pro-preview" to "gemini-3.1-pro-preview" across the codebase (468a0765)
- Refactor caching mechanisms to use Mutex instead of RwLock for improved concurrency (af7ebf32)
- Refactor sandbox configuration and enhance ANSI capabilities (cfb2f51a)
- Update default model to GPT-5.3 Codex and enhance terminal title management (9d3d7d38)
- Update default model to GPT-5.3 Codex, change provider to OpenAI, and enhance reasoning effort configuration (24d3af83)
- Remove xAI provider support and update default models in configuration files (ad6290aa)
- Refactor code for improved readability and consistency in function calls (91fdb43a)
#### Refactors
- Streamline task tracking logic and introduce TaskTrackingStatus for improved status management (3be2c784)
### Contributors
@vinhnx
## 0.82.6 - 2026-02-25
### Documentation
- Update TODO.md with critical release notes and improve terminal title (08d978a5) (@vinhnx)
- Update documentation map with enhanced topics and user questions (e7dc9369) (Vinh Nguyen)
### Features
- Implement dynamic terminal title updates based on agent activity (8c29b28f) (Vinh Nguyen)
### Other
- Implement XDG Base Directory Specification, add Zed extension file search commands, enhance PTY ANSI handling, and improve PTY pipe infrastructure. (03705867) (Vinh Nguyen)
- Update documentation links and paths across multiple files (76d81123) (Vinh Nguyen)
- Update LLM provider to Gemini and change default model and theme (9b20f88e) (Vinh Nguyen)
- Fix gemini streaming processor and update TODO.md (04927325) (Vinh Nguyen)
### Refactors
- Enhance changelog generation and update test script for versioning (4383577f) (Vinh Nguyen)
- Update documentation and improve terminology for unified search and code understanding (57e584eb) (Vinh Nguyen)
- Improve cacheable tool logic and enhance formatting in build script (2b276290) (Vinh Nguyen)
- Streamline provider resolution and update configuration to use Minimax (2b1d09c7) (Vinh Nguyen)
- Update Z.AI model constants and enhance parsing logic for legacy support (44e33c22) (Vinh Nguyen)
## 0.82.5 - 2026-02-25
### Bug Fixes
- Update versioning format to remove 'v' prefix in tags and URLs (081b6f73) (@vinhnx)
- Resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
- Correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- Prevent footer panic when hint is absent, refactor path argument to `&Path`, and optimize sidebar string truncation. (59525d4b) (@vinhnx)
- Update default model in configuration to glm-5:cloud (1700a7e4) (@vinhnx)
- Correct changelog generation to use the previous release tag instead of a fixed version. (b0437d27) (@vinhnx)
- Update chat input placeholders for clarity and improved user guidance (284a2390) (@vinhnx)
- Update chat input placeholders for clarity and improved user guidance (2e178d32) (@vinhnx)
- Disable scroll indicator in status bar (c5d4373d) (@vinhnx)
- Improve output messages for run completion and error handling (833564dd) (@vinhnguyen2308)
### Documentation
- Update documentation and improve clarity on execution plans, architectural invariants, and quality scores (27c61ef7) (@vinhnx)
- Update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- Update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- Add a guide for adding new models to AGENTS.md. (8648b584) (@vinhnx)
- Update TODO.md with additional PTY truncate display information and test references (cd3a3850) (@vinhnx)
- Update TODO.md with new tasks and references (049b6cae) (@vinhnx)
- Update TODO.md with examples and improve TUI display for truncated outputs (269395b1) (@vinhnx)
### Features
- Add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
- Add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- Add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- Implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- Implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- Add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- Add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- Implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- Add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- Implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- Add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- Implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- Add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- Add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- Refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (db99f4db) (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (c90f06e3) (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (83567152) (@vinhnx)
- Add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (d460c56d) (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (dc0742c0) (@vinhnx)
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (8b5b42a1) (@vinhnx)
- Add `prompt_cache_key` to OpenAI requests for improved cache locality and simplify Responses API usage logic. (45c9002e) (@vinhnx)
- Add top-level cache control to Anthropic requests, with TTL determined by breakpoint consumption. (91c0c9e4) (@vinhnx)
- Standardize MiniMax-M2.5 model identifier, promote it as the default, and update configuration defaults. (ff6dcef6) (@vinhnx)
- Introduce CI cost optimization strategies, add a new `--ci-only` release mode, and document release workflow details. (dd2f3168) (@vinhnx)
- Add prompt cache key to LLM requests and enhance unified_file tool execution diagnostics. (4073aed6) (@vinhnx)
- Refactor Ollama non-streaming request handling and add a fallback to non-streaming for initial stream failures. (30683331) (@vinhnx)
- Improve spooled tool output handling by verifying file existence and add a mechanism to suppress agent follow-up prompt detection for auto-generated prompts. (083ae71a) (@vinhnx)
- Enhance error handling and recovery mechanisms across various components (7902206c) (@vinhnx)
- Implement tool reentrancy guard to prevent recursive execution and improve panic reporting with `better-panic`. (44351bf5) (@vinhnx)
- Implement chunked reading for spooled tool outputs with improved agent messaging and update default LLM provider configuration. (b5b0c230) (@vinhnx)
- Add chunked file read spool progress tracking and refine token usage calculation for context management. (ef8f162d) (@vinhnx)
- Generate consolidated checksums.txt for releases and centralize script utilities into common.sh. (4bdb0808) (@vinhnx)
- Implement TaskTracker tool and enhance agent guards and documentation based on NL2Repo-Bench insights. (45a9a159) (@vinhnx)
- Integrate AI agent best practices into system prompts and loop detection for improved planning, root cause analysis, and uncertainty recognition. (91e5e295) (@vinhnx)
- Enhance documentation on grounding, uncertainty, and regression verification; improve loop detection guidance (064ea630) (@vinhnx)
- Enhance `AskUserChoice` with freeform input, custom labels, placeholders, and default selections. (53e0e111) (@vinhnx)
- Implement freeform text input for wizard modals, guided by system prompt and toggled by the Tab key. (9b54cdd6) (@vinhnx)
- Refine planning workflow transitions by adding more aliases, enabling contextual exit confirmations, and providing user guidance. (02bf1476) (@vinhnx)
- Set custom terminal title for VT Code TUI (9459f366) (@vinhnx)
- Migrate changelog generation to git-cliff and update related documentation (34d07804) (@vinhnx)
- Add support for additional tree-sitter languages and make them optional (b1c01414) (@vinhnguyen2308)
- Make tree-sitter language parsers optional and update dependencies to disable default features (5e268705) (@vinhnguyen2308)
### Other
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Update TODO.md (3391d6b9) (@vinhnx)
- Add spool chunk read tracking and enforce limits per turn (c09ef6d3) (@vinhnx)
- Increase max tool calls per turn to 48 and implement budget warning system in harness (7b6cade0) (@vinhnx)
- Rename UpdatePlanCommand to TaskTrackerCommand and refactor related files and documentation (b4520273) (@vinhnx)
- Add Planning workflow task tracker with CRUD functionality and integrate with existing tools (56c7e3b8) (@vinhnx)
- Add navigation loop guidance and improve planning workflow handling (5d162ae6) (@vinhnx)
- Refactor user input option generation and enhance markdown rendering in modals (a72a63c0) (@vinhnx)
- Add reference to git-cliff for changelog generation (8dd703f1) (@vinhnx)
- Refactor tool call handling and configuration limits (fcc783dd) (@vinhnx)
- Refactor registry and facade methods for improved error handling and readability; update configuration defaults and themes. (5ea7b7e1) (@vinhnx)
- Refactor error handling in various modules to improve resilience and logging; enhance cache management and concurrency controls. (035a14bb) (@vinhnx)
- Enhance TTY integration using crossterm's IsTty for consistent terminal handling (2a4e71d6) (@vinhnx)
- Implement unified exec terminal panel and enhance PTY session handling (f8505ba9) (@vinhnx)
- Refactor run command output handling and improve command summary rendering (59e8fab8) (@vinhnx)
- Implement plan artifact write handling and adjust LLM timeout logic (5edb5d66) (@vinhnx)
- Enhance timeout messages to include "continue" nudge for stalled turns in planning workflow (03cd0a65) (@vinhnx)
- Enhance planning workflow interview options with recommended choices for goals, constraints, and verification (3263981c) (@vinhnx)
- Refactor tool interaction: replace `ask_user_question` with `request_user_input`, enhance error classification, and improve function declaration caching. (2af17e60) (@vinhnx)
- Remove unified symbol extractor and related patterns; update language summarization to use file extensions; enhance README for clarity and feature descriptions; clean up unused imports in tools module. (aca4dbef) (@vinhnguyen2308)
### Refactors
- Remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
- Reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- Extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- Streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- Consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- Streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- Remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- Clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- Remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- Make TUI signal cleanup and dotfile permission backup UNIX-specific. (14a4f2d2) (@vinhnx)
- Improve previous SemVer tag identification by searching commit history in release scripts (0fcdce3d) (@vinhnx)
- Remove unified tool resolver module to streamline codebase (1384fe03) (@vinhnx)
- Remove unused TUI components and improve event handling for focus changes (99fa2620) (@vinhnx)
- Standardize continuation prefix handling in wrap_block_lines functions (0596e899) (@vinhnx)
## [unreleased]
### Bug Fixes
- Update versioning format to remove 'v' prefix in tags and URLs (@vinhnx)
- Resolve critical scrolling issue and remove unused slash command handlers (@vinhnx)
- Correct exec_code policy and update TODO for markdown rendering issue (@vinhnx)
- Update default model in configuration to glm-5:cloud (@vinhnx)
- Correct changelog generation to use the previous release tag instead of a fixed version. (@vinhnx)
- Update chat input placeholders for clarity and improved user guidance (Vinh Nguyen)
- Update chat input placeholders for clarity and improved user guidance (Vinh Nguyen)
- Disable scroll indicator in status bar (Vinh Nguyen)
### Documentation
- Update changelog for 0.79.2 [skip ci] (vtcode-release-bot)
- Update changelog for 0.79.3 [skip ci] (vtcode-release-bot)
- Update changelog for 0.79.4 [skip ci] (vtcode-release-bot)
- Update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (@vinhnx)
- Update changelog for 0.80.0 [skip ci] (vtcode-release-bot)
- Update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (@vinhnx)
- Update changelog for 0.80.1 [skip ci] (vtcode-release-bot)
- Update changelog for 0.81.0 [skip ci] (vtcode-release-bot)
- Update changelog for 0.81.1 [skip ci] (vtcode-release-bot)
- Update changelog for 0.81.2 [skip ci] (vtcode-release-bot)
- Update changelog for 0.81.3 [skip ci] (vtcode-release-bot)
- Add a guide for adding new models to AGENTS.md. (@vinhnx)
- Update changelog for 0.82.0 [skip ci] (vtcode-release-bot)
- Update changelog for 0.82.1 [skip ci] (vtcode-release-bot)
- Update TODO.md with additional PTY truncate display information and test references (@vinhnx)
- Update changelog for 0.82.2 [skip ci] (vtcode-release-bot)
- Update changelog for 0.83.0 [skip ci] (vtcode-release-bot)
- Update changelog for 0.82.3 [skip ci] (vtcode-release-bot)
- Update TODO.md with new tasks and references (Vinh Nguyen)
- Update TODO.md with examples and improve TUI display for truncated outputs (Vinh Nguyen)
### Features
- Add MiniMax M2.5 model support across various providers and update related constants (@vinhnx)
- Add Qwen3 Coder Next model support and update related constants (@vinhnx)
- Add skill bundle import/export functionality with zip support (@vinhnx)
- Implement planning workflow toggle and strip proposed plan blocks in rendering (@vinhnx)
- Implement in-process teammate runner and enhance team protocol messaging (@vinhnx)
- Add /share-log command to export session log as JSON for debugging (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (@vinhnx)
- Introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (@vinhnx)
- Add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (@vinhnx)
- Add sanitizer module for secret redaction and integrate into output handling (@vinhnx)
- Implement credential storage using OS keyring and file fallback (@vinhnx)
- Add timeout handling for turn metadata collection (@vinhnx)
- Implement mouse scroll handling for improved navigation (@vinhnx)
- Add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (@vinhnx)
- Implement secure storage for custom API keys using OS keyring (@vinhnx)
- Add CI workflows for building Linux and Windows binaries; optimize release process (@vinhnx)
- Add full CI mode to release script for all platforms (@vinhnx)
- Refactor build process to use conditional cross compilation for Linux and Windows (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (@vinhnx)
- Add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (@vinhnx)
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (@vinhnx)
- Add `prompt_cache_key` to OpenAI requests for improved cache locality and simplify Responses API usage logic. (@vinhnx)
- Add top-level cache control to Anthropic requests, with TTL determined by breakpoint consumption. (@vinhnx)
- Standardize MiniMax-M2.5 model identifier, promote it as the default, and update configuration defaults. (@vinhnx)
- Introduce CI cost optimization strategies, add a new `--ci-only` release mode, and document release workflow details. (@vinhnx)
- Add prompt cache key to LLM requests and enhance unified_file tool execution diagnostics. (@vinhnx)
- Refactor Ollama non-streaming request handling and add a fallback to non-streaming for initial stream failures. (@vinhnx)
- Improve spooled tool output handling by verifying file existence and add a mechanism to suppress agent follow-up prompt detection for auto-generated prompts. (@vinhnx)
- Enhance error handling and recovery mechanisms across various components (@vinhnx)
- Implement tool reentrancy guard to prevent recursive execution and improve panic reporting with `better-panic`. (@vinhnx)
- Implement chunked reading for spooled tool outputs with improved agent messaging and update default LLM provider configuration. (@vinhnx)
- Add chunked file read spool progress tracking and refine token usage calculation for context management. (@vinhnx)
- Generate consolidated checksums.txt for releases and centralize script utilities into common.sh. (@vinhnx)
- Implement TaskTracker tool and enhance agent guards and documentation based on NL2Repo-Bench insights. (@vinhnx)
- Integrate AI agent best practices into system prompts and loop detection for improved planning, root cause analysis, and uncertainty recognition. (@vinhnx)
- Enhance documentation on grounding, uncertainty, and regression verification; improve loop detection guidance (@vinhnx)
- Enhance `AskUserChoice` with freeform input, custom labels, placeholders, and default selections. (@vinhnx)
- Implement freeform text input for wizard modals, guided by system prompt and toggled by the Tab key. (@vinhnx)
- Refine planning workflow transitions by adding more aliases, enabling contextual exit confirmations, and providing user guidance. (Vinh Nguyen)
- Set custom terminal title for VT Code TUI (Vinh Nguyen)
- Migrate changelog generation to git-cliff and update related documentation (Vinh Nguyen)
### Other
- Prevent footer panic when hint is absent, refactor path argument to `&Path`, and optimize sidebar string truncation. (@vinhnx)
### Refactors
- Remove unused ReasoningSegment import from turn_processing.rs (@vinhnx)
- Reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (@vinhnx)
- Extract large event handler modules into smaller files for improved navigation (@vinhnx)
- Streamline file operations and enhance workspace path handling (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (@vinhnx)
- Replace manual file operations with shared utility functions for consistency (@vinhnx)
- Consolidate duplicated logic across workspace crates into shared utility functions (@vinhnx)
- Streamline MCP tool management and indexing in ToolRegistry (@vinhnx)
- Remove Rust cache step from release workflow (@vinhnx)
- Clean up code formatting and improve readability across multiple files (@vinhnx)
- Remove unused imports and enhance configuration for credential storage (@vinhnx)
- Make TUI signal cleanup and dotfile permission backup UNIX-specific. (@vinhnx)
- Improve previous SemVer tag identification by searching commit history in release scripts (@vinhnx)
- Remove unified tool resolver module to streamline codebase (Vinh Nguyen)
- Remove unused TUI components and improve event handling for focus changes (Vinh Nguyen)
- Standardize continuation prefix handling in wrap_block_lines functions (Vinh Nguyen)
## 0.82.3 - 2026-02-24
### Features
- Implement freeform text input for wizard modals, guided by system prompt and toggled by the Tab key. (9b54cdd6) (@vinhnx)
- Enhance `AskUserChoice` with freeform input, custom labels, placeholders, and default selections. (53e0e111) (@vinhnx)
- Enhance documentation on grounding, uncertainty, and regression verification; improve loop detection guidance (064ea630) (@vinhnx)
- Integrate AI agent best practices into system prompts and loop detection for improved planning, root cause analysis, and uncertainty recognition. (91e5e295) (@vinhnx)
- Implement TaskTracker tool and enhance agent guards and documentation based on NL2Repo-Bench insights. (45a9a159) (@vinhnx)
### Refactors
- improve previous SemVer tag identification by searching commit history in release scripts (0fcdce3d) (@vinhnx)
### Other
- Update commit (925f355d) (@vinhnx)
- Add reference to git-cliff for changelog generation (8dd703f1) (@vinhnx)
- Refactor user input option generation and enhance markdown rendering in modals (a72a63c0) (@vinhnx)
- Add navigation loop guidance and improve planning workflow handling (5d162ae6) (@vinhnx)
- Add Planning workflow task tracker with CRUD functionality and integrate with existing tools (56c7e3b8) (@vinhnx)
- Rename UpdatePlanCommand to TaskTrackerCommand and refactor related files and documentation (b4520273) (@vinhnx)
- Update commit (0642ff3b) (@vinhnx)
## 0.83.0 - 2026-02-24
### Features
- Implement freeform text input for wizard modals, guided by system prompt and toggled by the Tab key. (9b54cdd6) (@vinhnx)
- Enhance `AskUserChoice` with freeform input, custom labels, placeholders, and default selections. (53e0e111) (@vinhnx)
- Enhance documentation on grounding, uncertainty, and regression verification; improve loop detection guidance (064ea630) (@vinhnx)
- Integrate AI agent best practices into system prompts and loop detection for improved planning, root cause analysis, and uncertainty recognition. (91e5e295) (@vinhnx)
- Implement TaskTracker tool and enhance agent guards and documentation based on NL2Repo-Bench insights. (45a9a159) (@vinhnx)
### Refactors
- improve previous SemVer tag identification by searching commit history in release scripts (0fcdce3d) (@vinhnx)
### Other
- Refactor user input option generation and enhance markdown rendering in modals (a72a63c0) (@vinhnx)
- Add navigation loop guidance and improve planning workflow handling (5d162ae6) (@vinhnx)
- Add Planning workflow task tracker with CRUD functionality and integrate with existing tools (56c7e3b8) (@vinhnx)
- Rename UpdatePlanCommand to TaskTrackerCommand and refactor related files and documentation (b4520273) (@vinhnx)
- Update commit (0642ff3b) (@vinhnx)
## 0.82.2 - 2026-02-23
### Features
- Add chunked file read spool progress tracking and refine token usage calculation for context management. (ef8f162d) (@vinhnx)
- Implement chunked reading for spooled tool outputs with improved agent messaging and update default LLM provider configuration. (b5b0c230) (@vinhnx)
- Implement tool reentrancy guard to prevent recursive execution and improve panic reporting with `better-panic`. (44351bf5) (@vinhnx)
- enhance error handling and recovery mechanisms across various components (7902206c) (@vinhnx)
- improve spooled tool output handling by verifying file existence and add a mechanism to suppress agent follow-up prompt detection for auto-generated prompts. (083ae71a) (@vinhnx)
- refactor Ollama non-streaming request handling and add a fallback to non-streaming for initial stream failures. (30683331) (@vinhnx)
- Add prompt cache key to LLM requests and enhance unified_file tool execution diagnostics. (4073aed6) (@vinhnx)
- introduce CI cost optimization strategies, add a new `--ci-only` release mode, and document release workflow details. (dd2f3168) (@vinhnx)
- standardize MiniMax-M2.5 model identifier, promote it as the default, and update configuration defaults. (ff6dcef6) (@vinhnx)
- Add top-level cache control to Anthropic requests, with TTL determined by breakpoint consumption. (91c0c9e4) (@vinhnx)
- Add `prompt_cache_key` to OpenAI requests for improved cache locality and simplify Responses API usage logic. (45c9002e) (@vinhnx)
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (8b5b42a1) (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (dc0742c0) (@vinhnx)
- add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (d460c56d) (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (83567152) (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (c90f06e3) (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (db99f4db) (@vinhnx)
- refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct changelog generation to use the previous release tag instead of a fixed version. (b0437d27) (@vinhnx)
- Update default model in configuration to glm-5:cloud (1700a7e4) (@vinhnx)
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- Make TUI signal cleanup and dotfile permission backup UNIX-specific. (14a4f2d2) (@vinhnx)
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- Update TODO.md with additional PTY truncate display information and test references (cd3a3850) (@vinhnx)
- Add a guide for adding new models to AGENTS.md. (8648b584) (@vinhnx)
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (cf5d5f7e) (@vinhnx)
- Release (28a8476b) (@vinhnx)
- Release (77422bee) (@vinhnx)
- Release (14027810) (@vinhnx)
- Release (6c982a35) (@vinhnx)
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Increase max tool calls per turn to 48 and implement budget warning system in harness (7b6cade0) (@vinhnx)
- Add spool chunk read tracking and enforce limits per turn (c09ef6d3) (@vinhnx)
- Fix: Prevent footer panic when hint is absent, refactor path argument to `&Path`, and optimize sidebar string truncation. (59525d4b) (@vinhnx)
- Update TODO.md (3391d6b9) (@1097578+vinhnx)
- Update commit (9d05e9de) (@vinhnx)
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.82.1 - 2026-02-20
### Features
- introduce CI cost optimization strategies, add a new `--ci-only` release mode, and document release workflow details. (dd2f3168) (@vinhnx)
- standardize MiniMax-M2.5 model identifier, promote it as the default, and update configuration defaults. (ff6dcef6) (@vinhnx)
- Add top-level cache control to Anthropic requests, with TTL determined by breakpoint consumption. (91c0c9e4) (@vinhnx)
- Add `prompt_cache_key` to OpenAI requests for improved cache locality and simplify Responses API usage logic. (45c9002e) (@vinhnx)
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (8b5b42a1) (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (dc0742c0) (@vinhnx)
- add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (d460c56d) (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (83567152) (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (c90f06e3) (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (db99f4db) (@vinhnx)
- refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- Make TUI signal cleanup and dotfile permission backup UNIX-specific. (14a4f2d2) (@vinhnx)
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- Add a guide for adding new models to AGENTS.md. (8648b584) (@vinhnx)
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (28a8476b) (@vinhnx)
- Release (77422bee) (@vinhnx)
- Release (14027810) (@vinhnx)
- Release (6c982a35) (@vinhnx)
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Update commit (9d05e9de) (@vinhnx)
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.82.0 - 2026-02-20
### Features
- Implement Gemini prompt caching with TTL using a new `CacheControl` part and add support for Gemini 3.1 Pro preview models. (8b5b42a1) (@vinhnx)
- Implement Gemini 3.1 Pro Preview models with updated token limits and system prompt handling. (dc0742c0) (@vinhnx)
- add Claude Sonnet 4.6 model support and integrate it across model definitions, parsing, catalog, and documentation. (d460c56d) (@vinhnx)
- Implement mouse text selection in the TUI and add a new `vtcode.toml` configuration file. (83567152) (@vinhnx)
- Render GFM tables inside markdown code blocks as tables and prevent word-wrapping for table lines in the TUI. (c90f06e3) (@vinhnx)
- Implement mouse scroll support for TUI session and history picker, and update default agent configuration to Ollama. (db99f4db) (@vinhnx)
- refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- Add a guide for adding new models to AGENTS.md. (8648b584) (@vinhnx)
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (77422bee) (@vinhnx)
- Release (14027810) (@vinhnx)
- Release (6c982a35) (@vinhnx)
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Update commit (9d05e9de) (@vinhnx)
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.81.3 - 2026-02-20
### Features
- refactor build process to use conditional cross compilation for Linux and Windows (d15bb558) (@vinhnx)
- add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (14027810) (@vinhnx)
- Release (6c982a35) (@vinhnx)
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.81.2 - 2026-02-19
### Features
- add full CI mode to release script for all platforms (326a2c8c) (@vinhnx)
- add CI workflows for building Linux and Windows binaries; optimize release process (090bebb4) (@vinhnx)
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (6c982a35) (@vinhnx)
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.81.1 - 2026-02-17
### Features
- implement secure storage for custom API keys using OS keyring (3da5a60a) (@vinhnx)
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- clean up configuration file by removing unused custom API keys and simplifying array formatting (ad594e41) (@vinhnx)
- Release (4bd94bdb) (@vinhnx)
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.81.0 - 2026-02-16
### Features
- add Qwen3.5-397B-A17B model with hybrid architecture and update configuration (26a9a7ee) (@vinhnx)
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (800c7069) (@vinhnx)
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.80.1 - 2026-02-16
### Features
- implement mouse scroll handling for improved navigation (24a2d640) (@vinhnx)
- add timeout handling for turn metadata collection (1b1f91d4) (@vinhnx)
- implement credential storage using OS keyring and file fallback (1e94c71a) (@vinhnx)
- add sanitizer module for secret redaction and integrate into output handling (4263808d) (@vinhnx)
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Bug Fixes
- correct exec_code policy and update TODO for markdown rendering issue (9b87f88b) (@vinhnx)
- resolve critical scrolling issue and remove unused slash command handlers (bcb81434) (@vinhnx)
### Refactors
- remove unused imports and enhance configuration for credential storage (b79f2bd2) (@vinhnx)
- clean up code formatting and improve readability across multiple files (b1ae3ad9) (@vinhnx)
- remove Rust cache step from release workflow (e36c5f55) (@vinhnx)
- streamline MCP tool management and indexing in ToolRegistry (a5c3677b) (@vinhnx)
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update documentation for TECH_DEBT_TRACKER and QUALITY_SCORE; add tests for subagent loading and file operations (fb7c0944) (@vinhnx)
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (a08f765d) (@vinhnx)
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Enhance model behavior configuration for LLM providers (0fa12334) (@vinhnx)
- Add scripts for documentation link validation and markdown location checks (7290fc5b) (@vinhnx)
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.80.0 - 2026-02-16
### Features
- add Kimi K2.5 model support across OpenRouter, Ollama, and HuggingFace. (fddc4887) (@vinhnx)
- introduce agent legibility guidelines and refine steering message variants for clarity and structured output. (52b13dd1) (@vinhnx)
- Use configurable constants for agent session limits and expose the default max context tokens function. (21d1183f) (@vinhnx)
- Enhance reasoning display by introducing structured `ReasoningSegment` with stages and improved rendering in the TUI. (68d07c91) (@vinhnx)
- Implement agent steering mechanism to control runloop execution, including stop, pause, resume, and input injection capabilities. (ac806e0e) (@vinhnx)
- add /share-log command to export session log as JSON for debugging (64305820) (@vinhnx)
- implement in-process teammate runner and enhance team protocol messaging (4aff9318) (@vinhnx)
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Refactors
- consolidate duplicated logic across workspace crates into shared utility functions (a9df41fd) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (15c45f9a) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (7fcbe5f2) (@vinhnx)
- replace manual file operations with shared utility functions for consistency (54c447af) (@vinhnx)
- streamline file operations and enhance workspace path handling (f7ebb78d) (@vinhnx)
- extract large event handler modules into smaller files for improved navigation (9eda39e4) (@vinhnx)
- reimplement LLM streaming and event handling using AgentSessionController and its event sink mechanism. (95bcd08f) (@vinhnx)
- remove unused ReasoningSegment import from turn_processing.rs (9f4672d7) (@vinhnx)
### Documentation
- update contributing guidelines to reference CONTRIBUTING.md and AGENTS.md. (b4a1ef7a) (@vinhnx)
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (accdcc25) (@vinhnx)
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Update default model in vtcode.toml to minimax-m2:cloud; format code for consistency (e024a48d) (@vinhnx)
- Add additional_agent_dirs configuration option to vtcode.toml (0f6d1747) (@vinhnx)
- Remove custom prompts feature and related code; update error handling for agent actions; refine UI shortcuts and command handling (cfcc9765) (@vinhnx)
- Remove custom prompts feature and related code (dfc3ec5d) (@vinhnx)
- Update LLM provider configuration to use Ollama and remove unused typos.toml file (ca83bdcd) (@vinhnx)
- Enhance agent runner settings and improve loop detection; update CLI commands for reasoning effort and verbosity; refine patch handling in file operations; adjust markdown diff summary; modify configuration for LLM provider. (ff221d1d) (@vinhnx)
- Refactor agent runner and tool registry for improved error handling and normalization; enhance loop detection and add tests for new functionality. (101a2b07) (@vinhnx)
- Refactor LLM request handling, improve reasoning processing, and enhance tool command parsing (896f6c69) (@vinhnx)
- Implement streaming response rendering with reasoning support and helper functions (b3ba347b) (@vinhnx)
- Refactor file operations to use utility functions for directory creation and file reading/writing (7f800b3d) (@vinhnx)
- Implement unified session loop for agent execution and remove plugin marketplace commands (17d9597f) (@vinhnx)
- Add tests for model picker and prompt refinement functionality (0305390e) (@vinhnx)
- Enhance configuration and logging for agent and hooks; add mock MCP server for integration tests (59a48db6) (@vinhnx)
- Refactor agent runner to use AgentSessionState for session management and update related components (083fcac6) (@vinhnx)
- Refactor apply_patch handler for improved output handling; streamline sandbox policy mapping (67c0e05b) (@vinhnx)
- Refactor caching logic for improved performance; enhance error context handling; update tests for accurate cache statistics (5a10f0d7) (@vinhnx)
- Refactor exec_policy and command validation; consolidate HTTP client utilities; enhance error handling; improve file operations; update middleware to async; clean up validation cache; and adjust rate limiting implementation. (e6245af6) (@vinhnx)
- Adjust provider configurations and logging. (a649d023) (@vinhnx)
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.79.4 - 2026-02-14
### Features
- implement planning workflow toggle and strip proposed plan blocks in rendering (79f0327d) (@vinhnx)
- add skill bundle import/export functionality with zip support (dea5b5b7) (@vinhnx)
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Documentation
- update documentation and improve clarity on execution plans, architectural invariants, and quality scores feat: enhance system instruction generation to replace placeholders with unified tool guidance (27c61ef7) (@vinhnx)
### Chores
- Release (69df0e20) (@vinhnx)
- Release (10e4f284) (@vinhnx)
### Other
- Add support for inline skill bundles and network policies in skills (ddb9d280) (@vinhnx)
- Add architectural invariants, core beliefs, execution plans, quality score, and tech debt tracker documentation (80cf6d68) (@vinhnx)
## 0.79.3 - 2026-02-13
### Features
- add Qwen3 Coder Next model support and update related constants (5a4303e0) (@vinhnx)
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
### Chores
- Release (10e4f284) (@vinhnx)
## 0.79.2 - 2026-02-13
### Features
- add MiniMax M2.5 model support across various providers and update related constants (968963f1) (@vinhnx)
## v0.79.1 - 2026-02-13
### Features
- add support for MoonshotAI Kimi K2 models in ModelId (90e18ff2) (@vinhnx)
- complete model migration, fix test failures, and enhance UI stability (32f252ec) (@vinhnx)
- add pty_stream module and integrate it into tool pipeline execution (17b0c9d6) (@vinhnx)
### Refactors
- increase spooling thresholds and improve output handling for large tool outputs (dbae38d2) (@vinhnx)
- enhance file output handling and add no_spool flag for read operations (a3c134c4) (@vinhnx)
- enhance path validation logic and add lexical workspace check (605ea1ec) (@vinhnx)
- modularize tool output handling and enhance command safety validation (88146309) (@vinhnx)
- enhance tool validation and error messaging, modularize execution logic (2f6d20f0) (@vinhnx)
- streamline MCP event handling and enhance error content construction (9510ef79) (@vinhnx)
- remove dead code and streamline path handling functions (35289283) (@vinhnx)
- replace hardcoded Planning workflow strings with constants for consistency and maintainability (5d4d2407) (@vinhnx)
- update TODO with comprehensive code audit and optimization guidelines (4f96ff24) (@vinhnx)
- streamline error handling and validation logic; enhance retry safety checks (754ce484) (@vinhnx)
- enhance IDE context flushing and user confirmation handling; improve command auditing (b00d3b7f) (@vinhnx)
- improve tool validation and error handling; enhance test coverage for non-interactive environments (4b47208b) (@vinhnx)
- enhance error handling for tool arguments and improve rate limiting logic (386ce6aa) (@vinhnx)
- remove unused imports in turn_loop.rs for cleaner code (936e1885) (@vinhnx)
- improve code formatting and readability across multiple files (393f63e9) (@vinhnx)
- replace FxHashMap with LoopTracker for tool attempt tracking; optimize loop detection and history management (bb831f19) (@vinhnx)
- add token tracking validation in ContextManager; optimize turn balancer check intervals (4dd087c9) (@vinhnx)
- optimize tool signature handling and caching; enhance turn configuration extraction and prompt caching (b42c331a) (@vinhnx)
- comprehensive optimization of agent loop and tool execution pipeline (e97e91c0) (@vinhnx)
- improve code readability and structure across multiple files (2b5a2895) (@vinhnx)
- optimize line truncation logic in summarizers (4996c6ad) (@vinhnx)
- consolidate path resolution logic and remove redundant functions (d0a10bad) (@vinhnx)
### Tests
- skip TUI-dependent tests in non-interactive environments (4559d31e) (@vinhnx)
### Chores
- Release (4000421c) (@vinhnx)
### Other
- Update models and configurations for Gemini 3 and GLM-5; adjust tool capabilities and user confirmations (7f05b778) (@vinhnx)
- Add GLM-5 model support and remove deprecated GLM-4.5/4.6 models (29d0992a) (@vinhnx)
- Implement tool catalog state management and integrate into MCP tool lifecycle (18c73b54) (@vinhnx)
- Enhance planning workflow handling and tool safety validation; refactor prompt management and session loop logic (63a40249) (@vinhnx)
- Add prompt assembly mode and enhance tool validation (480ed33a) (@vinhnx)
- Add safety validation and transition functions for planning workflow handling (9db46b25) (@vinhnx)
- Refactor optimizer and tool result handling; enhance turn duration recording; update validation and state management; optimize loop detection; improve LLM request handling; remove fallback chains module; streamline tool execution checks; fix TUI modal search handling; adjust integration tests for tool usage. (9df107c3) (@vinhnx)
## v0.79.0 - 2026-02-13
### Features
- complete model migration, fix test failures, and enhance UI stability (32f252ec) (@vinhnx)
- add pty_stream module and integrate it into tool pipeline execution (17b0c9d6) (@vinhnx)
### Refactors
- increase spooling thresholds and improve output handling for large tool outputs (dbae38d2) (@vinhnx)
- enhance file output handling and add no_spool flag for read operations (a3c134c4) (@vinhnx)
- enhance path validation logic and add lexical workspace check (605ea1ec) (@vinhnx)
- modularize tool output handling and enhance command safety validation (88146309) (@vinhnx)
- enhance tool validation and error messaging, modularize execution logic (2f6d20f0) (@vinhnx)
- streamline MCP event handling and enhance error content construction (9510ef79) (@vinhnx)
- remove dead code and streamline path handling functions (35289283) (@vinhnx)
- replace hardcoded Planning workflow strings with constants for consistency and maintainability (5d4d2407) (@vinhnx)
- update TODO with comprehensive code audit and optimization guidelines (4f96ff24) (@vinhnx)
- streamline error handling and validation logic; enhance retry safety checks (754ce484) (@vinhnx)
- enhance IDE context flushing and user confirmation handling; improve command auditing (b00d3b7f) (@vinhnx)
- improve tool validation and error handling; enhance test coverage for non-interactive environments (4b47208b) (@vinhnx)
- enhance error handling for tool arguments and improve rate limiting logic (386ce6aa) (@vinhnx)
- remove unused imports in turn_loop.rs for cleaner code (936e1885) (@vinhnx)
- improve code formatting and readability across multiple files (393f63e9) (@vinhnx)
- replace FxHashMap with LoopTracker for tool attempt tracking; optimize loop detection and history management (bb831f19) (@vinhnx)
- add token tracking validation in ContextManager; optimize turn balancer check intervals (4dd087c9) (@vinhnx)
- optimize tool signature handling and caching; enhance turn configuration extraction and prompt caching (b42c331a) (@vinhnx)
- comprehensive optimization of agent loop and tool execution pipeline (e97e91c0) (@vinhnx)
- improve code readability and structure across multiple files (2b5a2895) (@vinhnx)
- optimize line truncation logic in summarizers (4996c6ad) (@vinhnx)
- consolidate path resolution logic and remove redundant functions (d0a10bad) (@vinhnx)
### Tests
- skip TUI-dependent tests in non-interactive environments (4559d31e) (@vinhnx)
### Other
- Update models and configurations for Gemini 3 and GLM-5; adjust tool capabilities and user confirmations (7f05b778) (@vinhnx)
- Add GLM-5 model support and remove deprecated GLM-4.5/4.6 models (29d0992a) (@vinhnx)
- Implement tool catalog state management and integrate into MCP tool lifecycle (18c73b54) (@vinhnx)
- Enhance planning workflow handling and tool safety validation; refactor prompt management and session loop logic (63a40249) (@vinhnx)
- Add prompt assembly mode and enhance tool validation (480ed33a) (@vinhnx)
- Add safety validation and transition functions for planning workflow handling (9db46b25) (@vinhnx)
- Refactor optimizer and tool result handling; enhance turn duration recording; update validation and state management; optimize loop detection; improve LLM request handling; remove fallback chains module; streamline tool execution checks; fix TUI modal search handling; adjust integration tests for tool usage. (9df107c3) (@vinhnx)
## v0.78.8 - 2026-02-09
### Refactors
- streamline release process and remove deprecated crate waiting logic (4436c3cc) (@vinhnx)
## v0.78.7 - 2026-02-09
*No significant changes*
## v0.78.6 - 2026-02-09
### Features
- add wait_for_crates_io function to ensure crate availability on crates.io (8b4ac577) (@vinhnx)
## v0.78.5 - 2026-02-09
### Other
- Remove outdated optimization notes and focus on DRY opportunities in the codebase (c9ac418b) (@vinhnx)
- Refactor to use rustc_hash::FxHashMap for improved performance and memory efficiency; update related structures and configurations. (3b166144) (@vinhnx)
## v0.78.4 - 2026-02-08
*No significant changes*
## v0.78.3 - 2026-02-08
### Features
- enhance crate publishing process with reliable version parsing and no-verify option (4fb5612c) (@vinhnx)
- implement delete_word_forward method in Session (5936d8bc) (@vinhnx)
- add delete_word_forward method to InputManager and update LayoutMode footer behavior (29e2dd09) (@vinhnx)
### Other
- Revert "feat: integrate `tui_input` crate for enhanced input management and modal search functionality" (b6e27465) (@vinhnx)
## v0.78.2 - 2026-02-08
### Features
- Display a scroll indicator in the TUI footer and adjust status height calculation based on layout mode. (6c5efc03) (@vinhnx)
- Introduce compile-time optimization guide and profiling script, and add general performance principles to TODO. (f1257ba9) (@vinhnx)
- add 'mono' theme and improve TUI modal search input handling. (7fd6334c) (@vinhnx)
- integrate `tui_input` crate for enhanced input management and modal search functionality (69a24ed3) (@vinhnx)
- improve plugin validation and enhance path resolution in PTY manager (a2ea12c1) (@vinhnx)
- add path utilities and normalize ASCII identifiers for improved path handling (5a888daf) (@vinhnx)
- implement command blocking during running tasks and update configuration for LLM provider (da55d7cd) (@vinhnx)
- enhance Planning workflow with reminders and execution prompts (ba69b139) (@vinhnx)
### Performance
- Cache session header lines and queued input previews to optimize TUI rendering performance and remove outdated content from TODO.md. (cf87bc80) (@vinhnx)
### Refactors
- Improve string truncation logic to ensure character boundaries are respected (83ebed35) (@vinhnx)
- Optimize I/O operations with buffered writes and simplify `ToolCallRecord`'s `tool_name` ownership. (a66a3a9d) (@vinhnx)
- remove scroll indicator from footer widget and associated UI logic. (4f060bed) (@vinhnx)
### Other
- Immprove (e3fc8d93) (@vinhnx)
- Add team context and teammate management features (b322e6cd) (@vinhnx)
## v0.78.1 - 2026-02-07
### Features
- enhance input handling with queue overlay and update input placeholders (8394895f) (@vinhnx)
- implement queue editing functionality and update input handling (6c0d373b) (@vinhnx)
- add support for inline data URLs and images in message content (90254442) (@vinhnx)
### Bug Fixes
- add finish_planning tool to planner agent and update tests (63a6835e) (@vinhnx)
### Other
- Implement collapsible pasted message handling and improve image path parsing (5efac36e) (@vinhnx)
## v0.78.0 - 2026-02-06
### Documentation
- add task summaries feature to agent teams and enhance subagent matching logic (5fa919c1) (@vinhnx)
### Other
- Refactor CLI argument documentation for clarity and consistency (087be239) (@vinhnx)
- Enhance user input tools to restrict usage to Planning workflow only (93675177) (@vinhnx)
- Improve output spooler and system prompt handling (255e0a44) (@vinhnx)
- Add experimental agent teams feature with slash commands and configuration (b70ce06e) (@vinhnx)
- Update script (35c8d01f) (@vinhnx)
- Remove code repetition and special casing of local providers (ec4b2099) (@gzsombor)
## v0.77.1 - 2026-02-06
### Chores
- Release (210e6503) (@vinhnx)
- update tool policy and improve tool registration descriptions (09e87101) (@vinhnx)
- update config - enable list_skills tool and adjust settings (401cc2f9) (@vinhnx)
### Other
- Update commit (717601ea) (@vinhnx)
- Add jq dependency check and improve cargo release process (5c9a2c82) (@vinhnx)
- Add support for effort parameter in Anthropic API and related validation (9519a720) (@vinhnx)
- Add adaptive thinking support for Claude Opus 4.6 model and update related configurations (19515dee) (@vinhnx)
- Add context management support to LLM requests and related components (e893cee1) (@vinhnx)
- Add support for Claude Opus 4.6 model with adaptive thinking and update related configurations (77d0d485) (@vinhnx)
- Implement tool safety checks, enhance wizard modal, and update configuration (574d60de) (@vinhnx)
- Add skills-ref commands for skill validation, listing, and prompt generation; update skill discovery paths and manifest structure (df2919d9) (@vinhnx)
- Enhance input widget styling with padding and background; update configuration theme and editing state (958f7c38) (@vinhnx)
- Cleanup TODO.md by removing outdated tasks and enhancing UI transition notes (21d1d5c6) (@vinhnx)
- Refactor spinner implementation and enhance cursor behavior during status updates (0b6828ac) (@vinhnx)
- Refactor test assertion for compact_title method in MCP event (4f44b68a) (@vinhnx)
- Implement command caching and gatekeeper policy; enhance file reading with async logging and performance tracking (e464ee31) (@vinhnx)
## v0.77.0 - 2026-02-06
### Chores
- update tool policy and improve tool registration descriptions (09e87101) (@vinhnx)
- update config - enable list_skills tool and adjust settings (401cc2f9) (@vinhnx)
### Other
- Add support for effort parameter in Anthropic API and related validation (9519a720) (@vinhnx)
- Add adaptive thinking support for Claude Opus 4.6 model and update related configurations (19515dee) (@vinhnx)
- Add context management support to LLM requests and related components (e893cee1) (@vinhnx)
- Add support for Claude Opus 4.6 model with adaptive thinking and update related configurations (77d0d485) (@vinhnx)
- Implement tool safety checks, enhance wizard modal, and update configuration (574d60de) (@vinhnx)
- Add skills-ref commands for skill validation, listing, and prompt generation; update skill discovery paths and manifest structure (df2919d9) (@vinhnx)
- Enhance input widget styling with padding and background; update configuration theme and editing state (958f7c38) (@vinhnx)
- Cleanup TODO.md by removing outdated tasks and enhancing UI transition notes (21d1d5c6) (@vinhnx)
- Refactor spinner implementation and enhance cursor behavior during status updates (0b6828ac) (@vinhnx)
- Refactor test assertion for compact_title method in MCP event (4f44b68a) (@vinhnx)
- Implement command caching and gatekeeper policy; enhance file reading with async logging and performance tracking (e464ee31) (@vinhnx)
## v0.76.2 - 2026-02-05
### Other
- Improve CI CD (79589790) (@vinhnx)
- Refactor MCP integration and update dependencies (b31b5407) (@vinhnx)
## v0.76.1 - 2026-02-05
### Other
- Refactor MCP integration and update dependencies (b31b5407) (@vinhnx)
## v0.76.0 - 2026-02-05
### Features
- add turn metadata support for LLM requests with git context (46a57d6d) (@vinhnx)
- enhance command safety checks for git subcommands and improve branch operation validation (ca9833f4) (@vinhnx)
- improve git changelog generator to group by commit types (9d2b46d1) (@vinhnx)
- implement shell snapshot feature to optimize command execution (e5d9d7fe) (@vinhnx)
- add git diff guidance to tool guidelines (934d723e) (@vinhnx)
- enhance agent message rendering with left padding and improved line handling (dc8025a6) (@vinhnx)
- update text deletion commands for improved line handling (7c003cae) (@vinhnx)
- clarify patch input parameters and remove 'diff' alias to prevent confusion (466549b7) (@vinhnx)
- enhance diff rendering with summary formatting and colorization (bbed557c) (@vinhnx)
- add support for inline streaming and recent spooled output retrieval (eee866c7) (@vinhnx)
- enhance diff view with changed lines count summary and line numbers (808464e9) (@vinhnx)
### Refactors
- simplify workspace directory creation in first run setup (6a35774d) (@vinhnx)
### Other
- Implement no_spool functionality for tool output and enhance cursor behavior during scrolling and shimmer states (477209fd) (@vinhnx)
- Refactor diff handling and rendering for improved clarity and summary display (f8dbf9e1) (@vinhnx)
## v0.75.2 - 2026-02-04
- Update TODO (fb065df0) (@vinhnx)
- feat: add Qwen3 Coder Next model with enhanced reasoning capabilities (26a65840) (@vinhnx)
- feat: enhance file output rendering to display diff content when applicable (0c05c762) (@vinhnx)
- feat: simplify debug script by removing sccache handling; enhance markdown diff rendering (cd8cdd28) (@vinhnx)
- feat: improve sccache error handling in debug script; retry without sccache on permission errors (29bff726) (@vinhnx)
- feat: enhance debug script to handle sccache permission errors during build and run (0beccb9e) (@vinhnx)
- feat: enhance message rendering for info boxes; group consecutive info messages and improve styling (3557bd25) (@vinhnx)
- feat: enhance UI styling and message rendering; improve error and info message handling (fa61cea9) (@vinhnx)
- feat: update tool policies, enhance message rendering, and modify default model configuration (436ac6cb) (@vinhnx)
- feat: reject hooks in skill definitions and update validation logic (6a4105e9) (@vinhnx)
- feat: update tool policies and enhance message handling; modify configuration for LLM provider (1c093ab7) (@vinhnx)
- Add webapp-testing skill with Playwright scripts and examples; introduce xlsx skill for spreadsheet handling (e2dfd86a) (@vinhnx)
- chore: update homebrew formula to v0.75.1 (b52a37dd)
## v0.75.1 - 2026-02-03
- refactor: reorganize release steps and update Homebrew process (e3d99f5a)
## v0.75.0 - 2026-02-03
- fix: resolve tool call ID mapping issue and update JSON handling in messages (9af9d34b) (@vinhnx)
- chore: update homebrew formula to v0.74.17 (2af3f3ff)
## v0.74.17 - 2026-02-03
- feat: add Step 3.5 Flash model and update configuration for OpenRouter (c50770ab) (@vinhnx)
- chore: update homebrew formula to v0.74.16 (a92e5a23)
## v0.74.16 - 2026-02-03
- Improve release (ff4ae644) (@vinhnx)
- chore(release): bump version to 0.74.15 [skip ci] (6a88018d) (@vinhnx)
- docs: update changelog for v0.74.15 [skip ci] (0b859919) (@vtcode-release-bot)
- Revert "refactor: enhance base URL resolution and improve JSON handling in request builder" (4cb8f2d9) (@vinhnx)
- chore: update homebrew formula to v0.74.14 (03234509)
## v0.74.15 - 2026-02-03
- Revert "refactor: enhance base URL resolution and improve JSON handling in request builder" (4cb8f2d9) (@vinhnx)
- chore: update homebrew formula to v0.74.14 (03234509)
## v0.74.14 - 2026-02-02
- chore(release): bump version to 0.74.13 [skip ci] (e49f412f) (@vinhnx)
- docs: update changelog for v0.74.13 [skip ci] (d64665d0) (@vtcode-release-bot)
- Update commit (171a2aa6) (@vinhnx)
- Refactor agent guidelines, improve spacing in TODO, and enhance model picker logic; update Anthropic provider tests and configuration (0dece6ac) (@vinhnx)
- refactor: enhance base URL resolution and improve JSON handling in request builder (584a82ca) (@vinhnx)
- docs: add behavioral guidelines to reduce common LLM coding mistakes (75ca745d) (@vinhnx)
- refactor: simplify conditional checks and remove unused imports (2602cbc8) (@vinhnx)
- refactor: update tool policies to allow all actions and improve terminal cleanup logic (0e54e2ff) (@vinhnx)
- chore: update homebrew formula to v0.74.12 (0e743b32)
## v0.74.13 - 2026-02-02
- Update commit (171a2aa6) (@vinhnx)
- Refactor agent guidelines, improve spacing in TODO, and enhance model picker logic; update Anthropic provider tests and configuration (0dece6ac) (@vinhnx)
- refactor: enhance base URL resolution and improve JSON handling in request builder (584a82ca) (@vinhnx)
- docs: add behavioral guidelines to reduce common LLM coding mistakes (75ca745d) (@vinhnx)
- refactor: simplify conditional checks and remove unused imports (2602cbc8) (@vinhnx)
- refactor: update tool policies to allow all actions and improve terminal cleanup logic (0e54e2ff) (@vinhnx)
- chore: update homebrew formula to v0.74.12 (0e743b32)
## v0.74.12 - 2026-02-02
- chore: update vtcode.gif resource (68cdf67b) (@vinhnx)
- docs: add compliance testing section and request object for Open Responses (ee2bcf14) (@vinhnx)
- chore(release): bump version to 0.74.11 [skip ci] (528d8846) (@vinhnx)
- docs: update changelog for v0.74.11 [skip ci] (dcdaef82) (@vtcode-release-bot)
- fix: update spinner finish behavior for cancellation handling (73a7d72f) (@vinhnx)
- refactor: remove unused set_defer_rendering method from StreamingReasoningState (e429daf0) (@vinhnx)
- refactor: remove deprecated model constants and clean up supported models list (eb9b6ff9) (@vinhnx)
- fix: resolve duplicate model entries and correct legacy model references (a712a191) (@vinhnx)
- Update model references to "claude-haiku-4-5" across configuration and tests (b06501f4) (@vinhnx)
- feat: add signal handling for graceful termination in TUI (f4de0101) (@vinhnx)
- Update model references from gpt-4 to gpt-5 across documentation and codebase (cec3d7c9) (@vinhnx)
- refactor: improve reasoning content comparison and suppress duplication in response rendering (87b066cf) (@vinhnx)
- chore: update default model and workspace trust settings in configuration fix: suppress duplicated content rendering in response handling refactor: clean up spacing logic in reasoning buffer add: implement streaming payload decoding helpers for OpenRouter (49a403fd) (@vinhnx)
- Fix release note (f7562c14) (@vinhnx)
- chore: update changelog header and release script title format (8f804b80) (@vinhnx)
- Duplicate badge links for Agent Skills and protocols (b9216314) (@1097578+vinhnx)
- Add star history section to README (3de92245) (@1097578+vinhnx)
- chore: update homebrew formula to v0.74.10 (c08a491c)
## v0.74.11 - 2026-02-02
- fix: update spinner finish behavior for cancellation handling (36d1f577) (@vinhnx)
- refactor: remove unused set_defer_rendering method from StreamingReasoningState (756c86ba) (@vinhnx)
- refactor: remove deprecated model constants and clean up supported models list (621b7373) (@vinhnx)
- fix: resolve duplicate model entries and correct legacy model references (7928cb44) (@vinhnx)
- Update model references to "claude-haiku-4-5" across configuration and tests (84a8d4ab) (@vinhnx)
- feat: add signal handling for graceful termination in TUI (a3ca378b) (@vinhnx)
- Update model references from gpt-4 to gpt-5 across documentation and codebase (2520e9fa) (@vinhnx)
- refactor: improve reasoning content comparison and suppress duplication in response rendering (bcce1fb1) (@vinhnx)
- chore: update default model and workspace trust settings in configuration fix: suppress duplicated content rendering in response handling refactor: clean up spacing logic in reasoning buffer add: implement streaming payload decoding helpers for OpenRouter (049663b0) (@vinhnx)
- Fix release note (d75a3f0d) (@vinhnx)
- chore: update changelog header and release script title format (744db41d) (@vinhnx)
- Add star history section to README (3de92245) (@1097578+vinhnx)
- chore: update homebrew formula to v0.74.10 (c08a491c)
## v0.74.10 - 2026-02-02
- Update commit (cbde5c0b)
- chore: update homebrew formula to v0.74.9 (6b604f22)
## v0.74.9 - 2026-02-02
### Documentation
##### [View changes on GitHub](https://github.com/vinhnx/vtcode/compare/v0.74.8...v0.74.9)
## v0.74.9 - 2026-02-02
*No significant changes*
##### [View changes on GitHub](https://github.com/vinhnx/vtcode/compare/v0.74.8...v0.74.9)
## v0.74.8 - 2026-02-02
### Refactors
- **commons**:
- **llm**:
### Documentation
##### [View changes on GitHub](https://github.com/vinhnx/vtcode/compare/v0.74.7...v0.74.8)
## v - 2026-02-02
### Refactors
- **commons**:
- **llm**:
##### [View changes on GitHub](https://github.com/vinhnx/vtcode/compare/v0.74.7...v)
## v0.74.7 - 2026-02-01
* Update commit (3edfdb95)
* fix: rename directory with colon to be Windows-compatible (f533addc)
* chore: update homebrew formula to v0.74.6 (dad20e9c)
## v0.74.6 - 2026-02-01
* Migrate LM Studio 0.4 REST API (4d12e993)
* Update release (bb930a4e)
* chore: update homebrew formula to v0.74.5 (dd7bdd0a)
## v0.74.5 - 2026-02-01
* Update CI (c9cf0a74)
* chore: update homebrew formula to v0.74.4 (1dbb0e27)
## v0.74.4 - 2026-02-01
* Fix: Skip hanging GitHub CLI refresh in build script (c77bb4aa)
* Fix: Skip hanging GitHub CLI refresh in release script (5bf61747)
* automation: add CI trigger and comprehensive release flow guide (d071618c)
* automation: add automatic gh auth switch and scope refresh (1b5f38ec)
* chore(release): bump version to 0.74.3 [skip ci] (6f365d16)
* docs: update changelog for v0.74.3 [skip ci] (d5b8110e)
* fix: skip gh auth checks during dry-run execution (8f4f8dae)
* automation: enhance release.sh with direct GitHub binary upload via gh CLI (9168134d)
* chore: add Windows cross-platform builds to release workflow (d497cf6d)
* feat: enhance AGENTS.md with new cargo commands and build performance tips (4df82383)
* feat: update imports in harness.rs and tests.rs for improved clarity (6e2b9432)
* chore: remove unnecessary blank line in run-tests.sh (c9422726)
* feat: update test commands to prefer cargo-nextest for faster execution (db00e155)
* feat: enhance Turn Diff Tracker with Agent Trace support and backward compatibility (a81074bd)
* feat: enhance Agent Trace support with async storage and serialization improvements (e777f259)
* Implement Agent Trace storage and specification for AI code attribution (d6748fc4)
* refactor: update process group management documentation for clarity (aff7157a)
* Implement process group management and graceful termination for child processes (d7aac98b)
* feat: implement wire API detection and version handling for Ollama (0cc27230)
* refactor: optimize development profile settings in Cargo.toml (cac65958)
* refactor: optimize development and test profiles in Cargo.toml (7f4cc397)
* fix: correct file path conversion in log_access method (ec7e3dc4)
* refactor: improve code formatting and readability across multiple files (d4ba430d)
* Fix badge links in README.md (e098b997)
* chore: update homebrew formula to v0.74.2 (744e7902)
## v0.74.3 - 2026-02-01
* fix: skip gh auth checks during dry-run execution (8f4f8dae)
* automation: enhance release.sh with direct GitHub binary upload via gh CLI (9168134d)
* chore: add Windows cross-platform builds to release workflow (d497cf6d)
* feat: enhance AGENTS.md with new cargo commands and build performance tips (4df82383)
* feat: update imports in harness.rs and tests.rs for improved clarity (6e2b9432)
* chore: remove unnecessary blank line in run-tests.sh (c9422726)
* feat: update test commands to prefer cargo-nextest for faster execution (db00e155)
* feat: enhance Turn Diff Tracker with Agent Trace support and backward compatibility (a81074bd)
* feat: enhance Agent Trace support with async storage and serialization improvements (e777f259)
* Implement Agent Trace storage and specification for AI code attribution (d6748fc4)
* refactor: update process group management documentation for clarity (aff7157a)
* Implement process group management and graceful termination for child processes (d7aac98b)
* feat: implement wire API detection and version handling for Ollama (0cc27230)
* refactor: optimize development profile settings in Cargo.toml (cac65958)
* refactor: optimize development and test profiles in Cargo.toml (7f4cc397)
* fix: correct file path conversion in log_access method (ec7e3dc4)
* refactor: improve code formatting and readability across multiple files (d4ba430d)
* Fix badge links in README.md (e098b997)
* chore: update homebrew formula to v0.74.2 (744e7902)
## v0.74.2 - 2026-01-31
* feat: add ACP authentication methods and configuration support (e51a5658)
* docs: update changelog for v0.74.2 [skip ci] (6d29ab5b)
* refactor: add dead code allowance for search_position and start_search method (39dc545b)
* Implement OpenRouter OAuth PKCE authentication flow and related utilities (c338e631)
* feat: add ANSI escape sequence parsing constants and improve handling in text utilities (d5641f54)
* refactor: remove unused Wrap widget import from history picker (67d5cc2c)
* feat: implement history picker for fuzzy command search (Ctrl+R) (e529e7a0)
* refactor: reorder use statements for clarity in theme module (63495bea)
* Update README.md to fix badge links (7b44cfa4)
* chore: update homebrew formula to v0.74.1 (73b0bf06)
## v0.74.2 - 2026-01-31
* refactor: add dead code allowance for search_position and start_search method (39dc545b)
* Implement OpenRouter OAuth PKCE authentication flow and related utilities (c338e631)
* feat: add ANSI escape sequence parsing constants and improve handling in text utilities (d5641f54)
* refactor: remove unused Wrap widget import from history picker (67d5cc2c)
* feat: implement history picker for fuzzy command search (Ctrl+R) (e529e7a0)
* refactor: reorder use statements for clarity in theme module (63495bea)
* Update README.md to fix badge links (7b44cfa4)
* chore: update homebrew formula to v0.74.1 (73b0bf06)
## v0.74.1 - 2026-01-31
* refactor: update terminal theme to ciapre-dark (157b61cd)
* refactor: improve formatting of model pull commands in Ollama provider documentation (252114fc)
* refactor: add Kimi K2.5 and GLM 4.7 models to Ollama provider documentation and tests (84b6c722)
* refactor: add Kimi K2.5 model support and update related configurations (6e83202d)
* refactor: improve table formatting in color guidelines documentation (d33b7afa)
* refactor: implement color accessibility features and update configuration options (6422c103)
* refactor: enhance cursor visibility logic by adding status spinner check (910506f9)
* refactor: update malloc warning suppression in debug script (839dbd68)
* refactor: improve malloc warning suppression and enhance spinner behavior in UI interactions (241c53fe)
* refactor: enhance local build process for macOS and Linux in release scripts (98cb01fb)
## v0.74.0 - 2026-01-31
* Fix permission (e2511d7e)
* docs: update AGENTS.md for improved clarity and formatting (775a0db7)
* refactor: update logging in TUI code to use tracing instead of println and eprintln (3393f2e0)
* refactor: replace eprintln with tracing for improved logging consistency (1d8f11d2)
* refactor: replace println with tracing for improved logging consistency (4ad8b586)
* fix: replace println with tracing debug for git repository check (c6c11053)
* feat: add skip_model_validation option to AnthropicConfig and update validation logic (f19fdb56)
* Refactor code for improved readability and consistency across multiple files (fec995fc)
* fix: preserve tool_exists when MCP tool check returns false (999f4a85)
* refactor: update tool execution methods and enhance context handling (e592f983)
* Refactor session loop and tool outcomes; remove unused code and improve context handling (016ef738)
* refactor: unify direct tool execution and expand interaction loop context with new tool-related services. (d7953097)
* refactor: refine `ToolOutcomeContext` lifetimes to improve mutable borrowing patterns and simplify context access. (a81254ef)
* refactor: adjust tool outcome context passing and borrowing in turn processing. (a197876a)
* refactor: centralize tool outcome handling parameters into a new `ToolOutcomeContext` struct. (73856006)
* Refactor tool call handling by centralizing execution, permission, and safety validation logic into dedicated outcome handlers and removing the execution module. (cfb8e1d7)
* refactor: extract metric recording and remove auto-exit planning workflow logic from tool execution result handling. (5e83a3a2)
* Refactor tool outcome handling in the agent runloop by introducing tool-specific retry limits, centralizing repetition tracking, and enhancing context conversion. (ce765244)
* Refactor tool outcome handling by consolidating success, failure, and timeout handlers, and updating tool repetition tracking to only count successful calls. (33a9b664)
* fix: enable `unified_file` tool in the sandbox and refactor diff preview styling to use a color palette. (15715b9e)
* feat: prevent duplicate LLM reasoning output and prioritize visible alias targets for hidden tools during lookup. (c434de61)
* feat: Implement dotfile protection with audit, backup, and guardian modules, and enhance tool registry alias resolution to prioritize LLM-visible tools. (62689a2c)
* feat: Implement session loading and controller switching, refreshing available commands on controller change and using constants for controller IDs. (9004549c)
* feat: Add collaboration controller switching tool and update `agent-client-protocol` dependency to 0.9.3, adapting API usage. (1bdf7a7f)
* feat: Introduce a standard Agent Client Protocol adapter and generalize ACP implementation details and tooling. (121184b5)
* fix: refine markdown styling logic for strong, heading, and inline code elements, and enhance theme-based accent application (04278e00)
* fix: update default theme, enable todo planning, refine tool output and display settings, and adjust tool policies for streamlined configuration (3308c57e)
* Update commit (db294299)
## v0.73.6 - 2026-01-30
* fix: update default theme, enable todo planning, refine tool output and display settings, and adjust tool policies for streamlined configuration (dcc7043e)
* fix: add persistence for editing and autonomous execution settings, and align theme with active configuration (ecacbc19)
* fix: add persistence for editing and autonomous execution settings, and align theme with active configuration (bfea2f52)
* fix: update default theme and reasoning effort, improve config overrides, adjust workspace trust mode, and enhance contribution docs (85c1ba7f)
* fix: add support for VTCODE_CONFIG_PATH, enhance configuration loading logic, and remove unused `.aiignore` file (bf6547aa)
* fix: improve shimmer animation handling, refactor spinner updates, and enhance status rendering logic (0a40ad46)
* fix: adjust tool policies, refactor text styling logic, and optimize message rendering indentation (8123ac34)
* fix: adjust color mappings, enhance markdown rendering logic, and add tests for new edge cases in tool policies and UI interactions (8309ffc1)
* fix: enforce tools_policy prompts, refactor workspace trust application, and enhance command safety checks (394aa59b)
* fix: enhance checksum verification logic across scripts, add fallback for individual sha256 files, and improve error handling (5abd7c60)
* fix: refactor ask command output handling, enhance pipeline detection, and centralize code extraction logic (b2aee725)
* fix: add spinner for long-running tasks, improve cursor handling, and streamline release fetching logic (484bfef1)
* fix: streamline platform-specific binary builds, refactor `ask` command implementation, and enhance local release workflow (d3bca9c2)
* chore: update homebrew formula to v0.73.5 (fd22abb9)
## v0.73.5 - 2026-01-29
* fix: improve release fetching with fallback for older versions, enhance platform-specific binary handling (5d1344a8)
* Improve deploy release (6b931dd7)
## v0.73.4 - 2026-01-29
* Fix vtcode-file-search build error (8437a13d)
* chore(release): bump version to 0.73.3 [skip ci] (48dd4c9b)
* docs: update changelog for v0.73.3 [skip ci] (d1c5e175)
* fix: streamline output handling in ask command and improve code extraction logic (8de09e16)
* chore: switch LLM provider to Ollama and update related configs, fix minor lint issues in release script (dc2e637d)
* chore: update homebrew formula to v0.73.2 (bf537280)
## v0.73.3 - 2026-01-29
* fix: streamline output handling in ask command and improve code extraction logic (8de09e16)
* chore: switch LLM provider to Ollama and update related configs, fix minor lint issues in release script (dc2e637d)
* chore: update homebrew formula to v0.73.2 (bf537280)
* fix: update GitHub release title format and improve changelog generation (97571aa2)
* chore: update homebrew formula to v0.73.1 (4ca75038)
## [Unreleased] - 2025-12-14
# [Version 0.73.1] - 2026-01-28
### Chores
- chore: update homebrew formula to v0.73.0 and fix update script
# [Version 0.73.0] - 2026-01-28
### Features
- feat: add GitHub Actions release workflow and update release script for better error handling
### Chores
- chore: enhance GitHub CLI authentication checks in release scripts
# [Version 0.72.4] - 2026-01-28
### Documentation
- docs: update changelog for v0.72.3 [skip ci]
- docs: update changelog for v0.73.0 [skip ci]
- docs: center align VT Code GIF in README
### Chores
- chore: fix README paths, benchmark inclusion, and release config
- chore(release): bump version to {{version}}
- chore: fix Cross.toml warnings and sync vtcode.toml version
- chore: update Cargo.toml to exclude resources directory and add VT Code GIF to README
- chore: update npm package.json to v0.72.2 version = [skip ci]
### Other Changes
- Fix docker build
- Refactor Cross.toml to consolidate Docker configuration for cross-compilation
- Enhance Open Responses specification conformance and update documentation
- Add technical whitepapers on security architecture and modular design principles
- Enhance Open Responses with sequenced events and improved item serialization
- Refactor code structure for improved readability and maintainability
- Implement Open Responses integration and configuration options
- Implement Open Responses specification with streaming events, output items, and response handling
# [Version 0.72.3] - 2026-01-28
### Documentation
- docs: update changelog for v0.73.0 [skip ci]
- docs: center align VT Code GIF in README
### Chores
- chore: fix Cross.toml warnings and sync vtcode.toml version
- chore: update Cargo.toml to exclude resources directory and add VT Code GIF to README
- chore: update npm package.json to v0.72.2 version = [skip ci]
### Other Changes
- Refactor Cross.toml to consolidate Docker configuration for cross-compilation
- Enhance Open Responses specification conformance and update documentation
- Add technical whitepapers on security architecture and modular design principles
- Enhance Open Responses with sequenced events and improved item serialization
- Refactor code structure for improved readability and maintainability
- Implement Open Responses integration and configuration options
- Implement Open Responses specification with streaming events, output items, and response handling
# [Version 0.73.0] - 2026-01-28
### Documentation
- docs: center align VT Code GIF in README
### Chores
- chore: update Cargo.toml to exclude resources directory and add VT Code GIF to README
- chore: update npm package.json to v0.72.2 version = [skip ci]
### Other Changes
- Refactor Cross.toml to consolidate Docker configuration for cross-compilation
- Enhance Open Responses specification conformance and update documentation
- Add technical whitepapers on security architecture and modular design principles
- Enhance Open Responses with sequenced events and improved item serialization
- Refactor code structure for improved readability and maintainability
- Implement Open Responses integration and configuration options
- Implement Open Responses specification with streaming events, output items, and response handling
# [Version 0.72.2] - 2026-01-28
### Documentation
- docs: update changelog for v0.72.1 [skip ci]
### Chores
- chore: update npm package.json to v0.72.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.72.0 version = [skip ci]
# [Version 0.72.1] - 2026-01-28
### Refactors
- refactor: enhance rendering logic for consistent tool output styling, simplify message spans, and update spinner handling for better readability
- refactor: remove DESIGN_SYSTEM.md, update default model to minimax-m2.5:cloud, and improve TUI spinner handling with `is_spinner_frame` function
- refactor: update authorship information across multiple crates, enhance TUI performance with increased tick rates, and integrate new tui-shimmer dependency for improved UI effects
- refactor: add follow-up prompts for truncated outputs and improve spooled file handling messages for enhanced user guidance
- refactor: enhance command status handling, improve loop detection logic, and update tool execution messages for clarity
- refactor: enhance rendering logic with dimming style, standardize long-running command locks, and update tooling policies for improved usability
- refactor: add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic
- refactor: add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic
- refactor: standardize color palette, update UI feedback styles, and improve markdown spacing configuration
- refactor: improve reasoning rendering logic, add deferred rendering support, and optimize duplicate content handling
- refactor: optimize markdown rendering with conditional line numbering, add diff language detection, and improve
- refactor: enhance line numbering in markdown code blocks, improve text trimming logic, and add support for "Reasoning" style rendering
- refactor: update default model to GPT-OSS, improve markdown rendering, and apply conditional/indentation optimizations
- refactor: switch default provider to Ollama, update model and API key configurations, and apply "if-let" refactoring for cleaner conditionals
### Documentation
- docs: update changelog for v0.72.0 [skip ci]
### Chores
- chore: update npm package.json to v0.72.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.7 version = [skip ci]
# [Version 0.72.0] - 2026-01-28
### Features
- feat: add new logo assets and update existing images for branding consistency
### Refactors
- refactor: enhance rendering logic for consistent tool output styling, simplify message spans, and update spinner handling for better readability
- refactor: remove DESIGN_SYSTEM.md, update default model to minimax-m2.5:cloud, and improve TUI spinner handling with `is_spinner_frame` function
- refactor: update authorship information across multiple crates, enhance TUI performance with increased tick rates, and integrate new tui-shimmer dependency for improved UI effects
- refactor: add follow-up prompts for truncated outputs and improve spooled file handling messages for enhanced user guidance
- refactor: enhance command status handling, improve loop detection logic, and update tool execution messages for clarity
- refactor: enhance rendering logic with dimming style, standardize long-running command locks, and update tooling policies for improved usability
- refactor: add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic
- refactor: add cargo command serialization to prevent file lock contention, improve PTY tool timeout handling, and enhance error recovery logic
- refactor: standardize color palette, update UI feedback styles, and improve markdown spacing configuration
- refactor: improve reasoning rendering logic, add deferred rendering support, and optimize duplicate content handling
- refactor: optimize markdown rendering with conditional line numbering, add diff language detection, and improve
- refactor: enhance line numbering in markdown code blocks, improve text trimming logic, and add support for "Reasoning" style rendering
- refactor: update default model to GPT-OSS, improve markdown rendering, and apply conditional/indentation optimizations
- refactor: switch default provider to Ollama, update model and API key configurations, and apply "if-let" refactoring for cleaner conditionals
- refactor: switch default provider to Hugging Face and add Moonshot Kimi K2.5 model support
- refactor: switch default LLM provider to Anthropics, enhance reasoning deduplication, and apply Codex-inspired output limits
- refactor: update default model and tool policy permissions, add Codex harness learnings documentation
### Documentation
- docs: update changelog for v0.71.7 [skip ci]
- docs: update changelog for v0.71.6 [skip ci]
- docs: update changelog for v0.71.5 [skip ci]
- docs: add comprehensive AgentSkills support section to README
### Chores
- chore: update npm package.json to v0.71.7 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: remove obsolete demo files and update README to reflect changes
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.4 version = [skip ci]
# [Version 0.71.7] - 2026-01-27
### Features
- feat: add new logo assets and update existing images for branding consistency
### Refactors
- refactor: switch default provider to Hugging Face and add Moonshot Kimi K2.5 model support
- refactor: switch default LLM provider to Anthropics, enhance reasoning deduplication, and apply Codex-inspired output limits
- refactor: update default model and tool policy permissions, add Codex harness learnings documentation
- refactor: update tool policy to allow file editing and enhance output spooling for PTY-related tools
### Documentation
- docs: update changelog for v0.71.6 [skip ci]
- docs: update changelog for v0.71.5 [skip ci]
- docs: add comprehensive AgentSkills support section to README
- docs: update changelog for v0.71.4 [skip ci]
### Chores
- chore: remove obsolete demo files and update README to reflect changes
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.4 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.3 version = [skip ci]
# [Version 0.71.6] - 2026-01-27
### Features
- feat: add new logo assets and update existing images for branding consistency
### Refactors
- refactor: switch default LLM provider to Anthropics, enhance reasoning deduplication, and apply Codex-inspired output limits
- refactor: update default model and tool policy permissions, add Codex harness learnings documentation
- refactor: update tool policy to allow file editing and enhance output spooling for PTY-related tools
### Documentation
- docs: update changelog for v0.71.5 [skip ci]
- docs: add comprehensive AgentSkills support section to README
- docs: update changelog for v0.71.4 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.4 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.3 version = [skip ci]
# [Version 0.71.5] - 2026-01-27
### Features
- feat: add new logo assets and update existing images for branding consistency
### Refactors
- refactor: switch default LLM provider to Anthropics, enhance reasoning deduplication, and apply Codex-inspired output limits
- refactor: update default model and tool policy permissions, add Codex harness learnings documentation
- refactor: update tool policy to allow file editing and enhance output spooling for PTY-related tools
### Documentation
- docs: add comprehensive AgentSkills support section to README
- docs: update changelog for v0.71.4 [skip ci]
### Chores
- chore: update npm package.json to v0.71.4 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.3 version = [skip ci]
# [Version 0.71.4] - 2026-01-27
### Refactors
- refactor: update tool policy to allow file editing and enhance output spooling for PTY-related tools
### Documentation
- docs: update changelog for v0.71.3 [skip ci]
### Chores
- chore: update npm package.json to v0.71.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.2 version = [skip ci]
# [Version 0.71.3] - 2026-01-26
### Documentation
- docs: update changelog for v0.71.2 [skip ci]
### Chores
- chore: update npm package.json to v0.71.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.71.1 version = [skip ci]
# [Version 0.71.2] - 2026-01-26
### Features
- feat: add UI support for modal layouts and wizard states in TUI session
- feat: add enhanced caching logic and tool execution pipeline improvements
- feat: streamline `file_ops` by removing legacy recursive search methods and enhance OpenAI provider with streaming logic
- feat: implement AgentRunner modularization for summarization, telemetry, tool access, and execution
- feat: modularize tool outcome handlers into separate files (failure, success, timeout, apply) and refactor implementation for better readability and maintainability
- feat: add `ZedAgent` implementation to support session management, tool execution, and client interaction
- feat: integrate `MCP client` with `ToolRegistry` and add functions for tool management
- feat: add `parse_openai_tool_calls` function to handle OpenAI tool call parsing logic
- feat: add OpenAI provider support for chat message parsing, request building, response parsing, and streaming decoder implementations
### Bug Fixes
- fix: remove `check_output.txt` to clean up outdated and obsolete error logs
- fix: address unresolved imports and modules in tests across multiple components
### Refactors
- refactor: adjust formatting, imports, and re-exports for improved consistency
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `text_tools.rs` to simplify codebase and eliminate unused functions
- refactor: remove unused LLM request structures and related configurations
- refactor: reorder imports across modules for consistency and readability
- refactor: reorder imports across modules for consistency and readability
- refactor: expand visibility for `parse_terminal_command` and `run_list_files` functions to improve module accessibility
- refactor: remove `models.rs` to simplify configuration and reduce redundancy in model management
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove obsolete `src/acp/zed.rs` file and related references to streamline the codebase
- refactor: remove `AnthropicProvider` and related implementations from the codebase to clean up unused functionality
- refactor: remove `OpenAIPromptCacheSettings` import from `xai.rs` to clean up unused dependencies
### Documentation
- docs: update changelog for v0.71.1 [skip ci]
- docs: update changelog for v0.71.0 [skip ci]
### Tests
- test: add environment variable handling and cleanup in `test_get_gemini_api_key_from_config`
### Chores
- chore: update npm package.json to v0.71.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.70.1 version = [skip ci]
# [Version 0.71.1] - 2026-01-26
### Features
- feat: add UI support for modal layouts and wizard states in TUI session
- feat: add enhanced caching logic and tool execution pipeline improvements
- feat: streamline `file_ops` by removing legacy recursive search methods and enhance OpenAI provider with streaming logic
- feat: implement AgentRunner modularization for summarization, telemetry, tool access, and execution
- feat: modularize tool outcome handlers into separate files (failure, success, timeout, apply) and refactor implementation for better readability and maintainability
- feat: add `ZedAgent` implementation to support session management, tool execution, and client interaction
- feat: integrate `MCP client` with `ToolRegistry` and add functions for tool management
- feat: add `parse_openai_tool_calls` function to handle OpenAI tool call parsing logic
- feat: add OpenAI provider support for chat message parsing, request building, response parsing, and streaming decoder implementations
- feat: add human-readable slug generator for plan file naming, update TUI header editing-state handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
### Bug Fixes
- fix: remove `check_output.txt` to clean up outdated and obsolete error logs
- fix: address unresolved imports and modules in tests across multiple components
### Refactors
- refactor: adjust formatting, imports, and re-exports for improved consistency
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `text_tools.rs` to simplify codebase and eliminate unused functions
- refactor: remove unused LLM request structures and related configurations
- refactor: reorder imports across modules for consistency and readability
- refactor: reorder imports across modules for consistency and readability
- refactor: expand visibility for `parse_terminal_command` and `run_list_files` functions to improve module accessibility
- refactor: remove `models.rs` to simplify configuration and reduce redundancy in model management
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove obsolete `src/acp/zed.rs` file and related references to streamline the codebase
- refactor: remove `AnthropicProvider` and related implementations from the codebase to clean up unused functionality
- refactor: remove `OpenAIPromptCacheSettings` import from `xai.rs` to clean up unused dependencies
- refactor: optimize memory usage and runtime efficiency, improve error handling, and enhance circuit breaker logic
### Documentation
- docs: update changelog for v0.71.0 [skip ci]
- docs: update changelog for v0.70.1 [skip ci]
### Tests
- test: add environment variable handling and cleanup in `test_get_gemini_api_key_from_config`
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.70.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.70.0 version = [skip ci]
# [Version 0.71.0] - 2026-01-26
### Features
- feat: add UI support for modal layouts and wizard states in TUI session
- feat: add enhanced caching logic and tool execution pipeline improvements
- feat: streamline `file_ops` by removing legacy recursive search methods and enhance OpenAI provider with streaming logic
- feat: implement AgentRunner modularization for summarization, telemetry, tool access, and execution
- feat: modularize tool outcome handlers into separate files (failure, success, timeout, apply) and refactor implementation for better readability and maintainability
- feat: add `ZedAgent` implementation to support session management, tool execution, and client interaction
- feat: integrate `MCP client` with `ToolRegistry` and add functions for tool management
- feat: add `parse_openai_tool_calls` function to handle OpenAI tool call parsing logic
- feat: add OpenAI provider support for chat message parsing, request building, response parsing, and streaming decoder implementations
- feat: add human-readable slug generator for plan file naming, update TUI header editing-state handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
### Bug Fixes
- fix: remove `check_output.txt` to clean up outdated and obsolete error logs
- fix: address unresolved imports and modules in tests across multiple components
### Refactors
- refactor: adjust formatting, imports, and re-exports for improved consistency
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `read_file_handler.rs`, `bash_runner.rs`, and unused code
- refactor: remove `text_tools.rs` to simplify codebase and eliminate unused functions
- refactor: remove unused LLM request structures and related configurations
- refactor: reorder imports across modules for consistency and readability
- refactor: reorder imports across modules for consistency and readability
- refactor: expand visibility for `parse_terminal_command` and `run_list_files` functions to improve module accessibility
- refactor: remove `models.rs` to simplify configuration and reduce redundancy in model management
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove Anthropic provider and OpenRouter implementation for codebase simplification
- refactor: remove obsolete `src/acp/zed.rs` file and related references to streamline the codebase
- refactor: remove `AnthropicProvider` and related implementations from the codebase to clean up unused functionality
- refactor: remove `OpenAIPromptCacheSettings` import from `xai.rs` to clean up unused dependencies
- refactor: optimize memory usage and runtime efficiency, improve error handling, and enhance circuit breaker logic
### Documentation
- docs: update changelog for v0.70.1 [skip ci]
### Tests
- test: add environment variable handling and cleanup in `test_get_gemini_api_key_from_config`
### Chores
- chore: update npm package.json to v0.70.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.70.0 version = [skip ci]
# [Version 0.70.1] - 2026-01-25
### Features
- feat: add human-readable slug generator for plan file naming, update TUI header editing-state handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
- feat: migrate `XAIProvider` to use the new `Responses API`, improve support for tools, caching, and error handling
- feat: introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation
- feat: introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation
- feat: refine tool output styling, and enhance agent configuration
- feat: refine tool output styling, and enhance agent configuration
- feat: add nested discovery for Claude skills, enhance SKILL.md parsing with default values, and update validation rules
- feat: add subagent system with optional enablement, commands, and configuration updates
### Refactors
- refactor: optimize memory usage and runtime efficiency, improve error handling, and enhance circuit breaker logic
- refactor: remove `ui.show_message_dividers` config, simplify divider logic, and enhance tool summary rendering
- refactor: transition `UnifiedCache` to use `RwLock` for interior mutability, enhance test coverage, and simplify cache operations
- refactor: disable subagents by default and update documentation with usage and configuration details
- refactor: remove unused fields and path handling from `HarnessEventEmitter` and `HarnessTurnState`
### Documentation
- docs: update changelog for v0.70.0 [skip ci]
### Chores
- chore: update npm package.json to v0.70.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.69.1 version = [skip ci]
# [Version 0.70.0] - 2026-01-24
### Features
- feat: introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation
- feat: introduce `InputHistoryEntry` to manage input with attachments, enhance reverse search and history navigation
- feat: refine tool output styling, and enhance agent configuration
- feat: refine tool output styling, and enhance agent configuration
- feat: add nested discovery for Claude skills, enhance SKILL.md parsing with default values, and update validation rules
- feat: add subagent system with optional enablement, commands, and configuration updates
- feat: add harness event emitter and session persistence for enhanced logging and state management
- feat: add adaptive logo SVGs for different color schemes
### Bug Fixes
- fix: update allowed tools list to include request_user_input
- fix: enhance output spooling logic for PTY commands and handle double-serialized JSON
### Refactors
- refactor: remove `ui.show_message_dividers` config, simplify divider logic, and enhance tool summary rendering
- refactor: transition `UnifiedCache` to use `RwLock` for interior mutability, enhance test coverage, and simplify cache operations
- refactor: disable subagents by default and update documentation with usage and configuration details
- refactor: remove unused fields and path handling from `HarnessEventEmitter` and `HarnessTurnState`
- refactor: update queue display to show follow-ups and improve styling
### Documentation
- docs: update changelog for v0.69.1 [skip ci]
- docs: update changelog for v0.69.0 [skip ci]
- docs: add note to check amp in vscode session
### Chores
- chore: update npm package.json to v0.69.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.68.2 version = [skip ci]
# [Version 0.69.1] - 2026-01-24
### Features
- feat: add harness event emitter and session persistence for enhanced logging and state management
- feat: add adaptive logo SVGs for different color schemes
- feat: implement batch file reading with token-efficient command transformation
- feat: add support for additional programming languages including swift in syntax highlighting and configuration
### Bug Fixes
- fix: update allowed tools list to include request_user_input
- fix: enhance output spooling logic for PTY commands and handle double-serialized JSON
### Refactors
- refactor: update queue display to show follow-ups and improve styling
- refactor: enhance token-efficient output handling and command parsing in executors
- refactor: update finish_planning policy to prompt; enhance OpenResponsesProvider with version handling and reasoning content support
### Documentation
- docs: update changelog for v0.69.0 [skip ci]
- docs: add note to check amp in vscode session
- docs: update changelog for v0.68.2 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.68.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.68.1 version = [skip ci]
# [Version 0.69.0] - 2026-01-24
### Features
- feat: add harness event emitter and session persistence for enhanced logging and state management
- feat: add adaptive logo SVGs for different color schemes
- feat: implement batch file reading with token-efficient command transformation
- feat: add support for additional programming languages including swift in syntax highlighting and configuration
### Bug Fixes
- fix: update allowed tools list to include request_user_input
- fix: enhance output spooling logic for PTY commands and handle double-serialized JSON
### Refactors
- refactor: update queue display to show follow-ups and improve styling
- refactor: enhance token-efficient output handling and command parsing in executors
- refactor: update finish_planning policy to prompt; enhance OpenResponsesProvider with version handling and reasoning content support
### Documentation
- docs: add note to check amp in vscode session
- docs: update changelog for v0.68.2 [skip ci]
### Chores
- chore: update npm package.json to v0.68.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.68.1 version = [skip ci]
# [Version 0.68.2] - 2026-01-24
### Features
- feat: implement batch file reading with token-efficient command transformation
- feat: add support for additional programming languages including swift in syntax highlighting and configuration
### Bug Fixes
- fix: remove unused methods is_planner_active and is_coder_active
### Refactors
- refactor: enhance token-efficient output handling and command parsing in executors
- refactor: update finish_planning policy to prompt; enhance OpenResponsesProvider with version handling and reasoning content support
- refactor: increase max_conversation_turns to 150 and streamline allowed_tools format; add auto_finish_planning_attempted to context for improved planning workflow handling
- refactor: integrate clean_reasoning_text function to streamline reasoning text handling across multiple modules
- refactor: rename default method to default_cache and update default implementations for various structs
### Documentation
- docs: update changelog for v0.68.1 [skip ci]
### Chores
- chore: update npm package.json to v0.68.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: clean up Cargo.toml and Cargo.lock by removing unused dependencies and updating package versions
- chore: update Rust version to 1.93.0 in documentation and templates
- chore: update npm package.json to v0.68.0 version = [skip ci]
# [Version 0.68.1] - 2026-01-23
### Bug Fixes
- fix: remove unused methods is_planner_active and is_coder_active
### Refactors
- refactor: increase max_conversation_turns to 150 and streamline allowed_tools format; add auto_finish_planning_attempted to context for improved planning workflow handling
- refactor: integrate clean_reasoning_text function to streamline reasoning text handling across multiple modules
- refactor: rename default method to default_cache and update default implementations for various structs
### Documentation
- docs: update changelog for v0.68.0 [skip ci]
### Chores
- chore: clean up Cargo.toml and Cargo.lock by removing unused dependencies and updating package versions
- chore: update Rust version to 1.93.0 in documentation and templates
- chore: update npm package.json to v0.68.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.67.0 version = [skip ci]
# [Version 0.68.0] - 2026-01-22
### Refactors
- refactor: simplify codebase by auditing markdown.rs, removing unused tests, and standardizing effort parameter in tool calls
- refactor: remove tui-syntax-highlight dependency and streamline syntax highlighting implementation
- refactor: adjust output thresholds and preview line counts for improved token efficiency
### Documentation
- docs: update changelog for v0.67.0 [skip ci]
### Chores
- chore: update npm package.json to v0.67.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.8 version = [skip ci]
# [Version 0.67.0] - 2026-01-22
### Features
- feat: enhance CLI with quick start guidance and slash command notes
### Refactors
- refactor: simplify codebase by auditing markdown.rs, removing unused tests, and standardizing effort parameter in tool calls
- refactor: remove tui-syntax-highlight dependency and streamline syntax highlighting implementation
- refactor: adjust output thresholds and preview line counts for improved token efficiency
### Documentation
- docs: update changelog for v0.66.8 [skip ci]
### Chores
- chore: update npm package.json to v0.66.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.7 version = [skip ci]
# [Version 0.66.8] - 2026-01-22
### Features
- feat: enhance CLI with quick start guidance and slash command notes
### Refactors
- refactor: use AsRef trait for string conversion in command rendering
### Documentation
- docs: update changelog for v0.66.7 [skip ci]
### Chores
- chore: update npm package.json to v0.66.7 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.6 version = [skip ci]
# [Version 0.66.7] - 2026-01-22
### Bug Fixes
- fix: resolve ambiguous AsRef trait for Cow in zed.rs
- fix: resolve ambiguous AsRef trait for Cow<'_, str>
### Refactors
- refactor: use AsRef trait for string conversion in command rendering
### Documentation
- docs: update changelog for v0.66.6 [skip ci]
- docs: update changelog for v0.66.5 [skip ci]
- docs: update changelog for v0.66.4 [skip ci]
### Chores
- chore: update npm package.json to v0.66.6 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.3 version = [skip ci]
# [Version 0.66.6] - 2026-01-22
### Bug Fixes
- fix: resolve ambiguous AsRef trait for Cow in zed.rs
- fix: resolve ambiguous AsRef trait for Cow<'_, str>
### Refactors
- refactor: optimize string handling with dereferencing in multiple files
### Documentation
- docs: update changelog for v0.66.5 [skip ci]
- docs: update changelog for v0.66.4 [skip ci]
- docs: update changelog for v0.66.3 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.2 version = [skip ci]
# [Version 0.66.5] - 2026-01-22
### Bug Fixes
- fix: resolve ambiguous AsRef trait for Cow in zed.rs
- fix: resolve ambiguous AsRef trait for Cow<'_, str>
### Refactors
- refactor: optimize string handling with dereferencing in multiple files
### Documentation
- docs: update changelog for v0.66.4 [skip ci]
- docs: update changelog for v0.66.3 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.2 version = [skip ci]
# [Version 0.66.4] - 2026-01-22
### Bug Fixes
- fix: resolve ambiguous AsRef trait for Cow<'_, str>
### Refactors
- refactor: optimize string handling with dereferencing in multiple files
### Documentation
- docs: update changelog for v0.66.3 [skip ci]
### Chores
- chore: update npm package.json to v0.66.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.66.2 version = [skip ci]
# [Version 0.66.3] - 2026-01-22
### Features
- feat: add support for image URLs in @ pattern parsing and implement vision support for LLM providers
### Refactors
- refactor: optimize string handling with dereferencing in multiple files
- refactor: enhance log event filtering and improve user message styling in TUI
- refactor: optimize inline event handling and improve command safety checks
- refactor: streamline syntax highlighting by introducing a dedicated module and optimizing theme management
- refactor: update thinking budget constants and enhance extended thinking configuration
- refactor: implement Chain-of-Thought monitoring and context anxiety management patterns
- refactor: improve history navigation and update input handling
- refactor: enhance history navigation and update inline event handling
- refactor: update test assertions for clarity and accuracy
- refactor: update TODO list with improved queue messages UI and handling
- refactor: update path parameter types from PathBuf to Path for consistency
- refactor: simplify conditional checks for context awareness in prompt building
### Documentation
- docs: update changelog for v0.66.2 [skip ci]
- docs: update changelog for v0.67.0 [skip ci]
- docs: update changelog for v0.67.0 [skip ci]
### Tests
- test: add streaming event deserialization tests
### Chores
- chore: update npm package.json to v0.66.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: add #[allow(dead_code)] annotations to unused items across multiple files
- chore: update npm package.json to v0.66.1 version = [skip ci]
# [Version 0.66.2] - 2026-01-22
### Features
- feat: add support for image URLs in @ pattern parsing and implement vision support for LLM providers
- feat: Add Z.AI GLM-4.7-Flash model support and update configuration
- feat: Add extended thinking configuration for Anthropic models
- feat: Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option.
### Refactors
- refactor: enhance log event filtering and improve user message styling in TUI
- refactor: optimize inline event handling and improve command safety checks
- refactor: streamline syntax highlighting by introducing a dedicated module and optimizing theme management
- refactor: update thinking budget constants and enhance extended thinking configuration
- refactor: implement Chain-of-Thought monitoring and context anxiety management patterns
- refactor: improve history navigation and update input handling
- refactor: enhance history navigation and update inline event handling
- refactor: update test assertions for clarity and accuracy
- refactor: update TODO list with improved queue messages UI and handling
- refactor: update path parameter types from PathBuf to Path for consistency
- refactor: simplify conditional checks for context awareness in prompt building
### Documentation
- docs: update changelog for v0.67.0 [skip ci]
- docs: update changelog for v0.67.0 [skip ci]
- docs: update changelog for v0.66.1 [skip ci]
- docs: update changelog for v0.66.0 [skip ci]
### Tests
- test: add streaming event deserialization tests
### Chores
- chore: add #[allow(dead_code)] annotations to unused items across multiple files
- chore: update npm package.json to v0.66.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
# [Version 0.67.0] - 2026-01-20
### Features
- feat: add support for image URLs in @ pattern parsing and implement vision support for LLM providers
- feat: Add Z.AI GLM-4.7-Flash model support and update configuration
- feat: Add extended thinking configuration for Anthropic models
- feat: Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option.
### Refactors
- refactor: update thinking budget constants and enhance extended thinking configuration
- refactor: implement Chain-of-Thought monitoring and context anxiety management patterns
- refactor: improve history navigation and update input handling
- refactor: enhance history navigation and update inline event handling
- refactor: update test assertions for clarity and accuracy
- refactor: update TODO list with improved queue messages UI and handling
- refactor: update path parameter types from PathBuf to Path for consistency
- refactor: simplify conditional checks for context awareness in prompt building
### Documentation
- docs: update changelog for v0.67.0 [skip ci]
- docs: update changelog for v0.66.1 [skip ci]
- docs: update changelog for v0.66.0 [skip ci]
### Tests
- test: add streaming event deserialization tests
### Chores
- chore: add #[allow(dead_code)] annotations to unused items across multiple files
- chore: update npm package.json to v0.66.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
# [Version 0.67.0] - 2026-01-20
### Features
- feat: add support for image URLs in @ pattern parsing and implement vision support for LLM providers
- feat: Add Z.AI GLM-4.7-Flash model support and update configuration
- feat: Add extended thinking configuration for Anthropic models
- feat: Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option.
### Refactors
- refactor: update thinking budget constants and enhance extended thinking configuration
- refactor: implement Chain-of-Thought monitoring and context anxiety management patterns
- refactor: improve history navigation and update input handling
- refactor: enhance history navigation and update inline event handling
- refactor: update test assertions for clarity and accuracy
- refactor: update TODO list with improved queue messages UI and handling
- refactor: update path parameter types from PathBuf to Path for consistency
- refactor: simplify conditional checks for context awareness in prompt building
### Documentation
- docs: update changelog for v0.66.1 [skip ci]
- docs: update changelog for v0.66.0 [skip ci]
### Tests
- test: add streaming event deserialization tests
### Chores
- chore: add #[allow(dead_code)] annotations to unused items across multiple files
- chore: update npm package.json to v0.66.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
# [Version 0.66.1] - 2026-01-19
### Features
- feat: Add Z.AI GLM-4.7-Flash model support and update configuration
- feat: Add extended thinking configuration for Anthropic models
- feat: Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option.
### Documentation
- docs: update changelog for v0.66.0 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.5 version = [skip ci]
# [Version 0.66.0] - 2026-01-19
### Features
- feat: Add Z.AI GLM-4.7-Flash model support and update configuration
- feat: Add extended thinking configuration for Anthropic models
- feat: Implement Anthropic token counting, allowing estimation of input tokens via a new configurable option.
- feat: add effort parameter for Claude Opus 4.5 to control token usage
### Documentation
- docs: update changelog for v0.65.5 [skip ci]
### Chores
- chore: update npm package.json to v0.65.5 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.4 version = [skip ci]
# [Version 0.65.5] - 2026-01-19
### Features
- feat: add effort parameter for Claude Opus 4.5 to control token usage
- feat: implement autonomous execution with reduced HITL prompts and update related configurations
- feat: enhance output spooling for read_file and unified_file with raw content extraction
- feat: enhance context awareness with token usage tracking and context window size
### Bug Fixes
- fix: prevent duplicate reasoning output during finalization
### Documentation
- docs: update changelog for v0.65.4 [skip ci]
### Chores
- chore: update npm package.json to v0.65.4 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.3 version = [skip ci]
# [Version 0.65.4] - 2026-01-18
### Features
- feat: implement autonomous execution with reduced HITL prompts and update related configurations
- feat: enhance output spooling for read_file and unified_file with raw content extraction
- feat: enhance context awareness with token usage tracking and context window size
- feat: Add max_conversation_turns configuration to various components and update tests
- feat: Implement autonomous loop detection with TUI warnings, integrate into agent runloop, and add project TODO documentation.
- feat: Enhance agent robustness with exponential backoff for circuit breakers, custom tool loop limits, and conversation turn limits.
- feat: Implement priority-based adaptive rate limiting, tiered cache eviction, and sliding window tool health tracking.
- feat: Update session limit messages to recommend persisting progress via artifacts like task.md/docs.
- feat: Dynamically configure conversation message and session turn limits, and remove telemetry from interaction loop parameters.
- feat: Implement adaptive rate limiting with priority-based scaling and integrate telemetry for tool usage tracking.
- feat: Enhance tool execution with circuit breakers, adaptive rate limiting, and health-based delegation, and introduce session telemetry and dynamic cache capacity management.
- feat: introduce dedicated modules for MCP lifecycle, slash command handling, and tool dispatch, and parallelize tool batch execution.
- feat: improve code block indentation normalization to handle mixed whitespace and refine markdown table rendering separators.
### Bug Fixes
- fix: prevent duplicate reasoning output during finalization
### Refactors
- refactor: Simplify tracing initialization with unwrap_or_default
- refactor: Replace map_or with is_none_or for improved clarity in MCP tool filtering
- refactor: Remove unnecessary cloning and assignment of `_updated_snapshot`.
- refactor: Inline table row rendering logic, remove duplicate parameters, and clean up the TODO list.
### Documentation
- docs: update changelog for v0.65.3 [skip ci]
### Chores
- chore: update npm package.json to v0.65.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.2 version = [skip ci]
# [Version 0.65.3] - 2026-01-18
### Features
- feat: Add max_conversation_turns configuration to various components and update tests
- feat: Implement autonomous loop detection with TUI warnings, integrate into agent runloop, and add project TODO documentation.
- feat: Enhance agent robustness with exponential backoff for circuit breakers, custom tool loop limits, and conversation turn limits.
- feat: Implement priority-based adaptive rate limiting, tiered cache eviction, and sliding window tool health tracking.
- feat: Update session limit messages to recommend persisting progress via artifacts like task.md/docs.
- feat: Dynamically configure conversation message and session turn limits, and remove telemetry from interaction loop parameters.
- feat: Implement adaptive rate limiting with priority-based scaling and integrate telemetry for tool usage tracking.
- feat: Enhance tool execution with circuit breakers, adaptive rate limiting, and health-based delegation, and introduce session telemetry and dynamic cache capacity management.
- feat: introduce dedicated modules for MCP lifecycle, slash command handling, and tool dispatch, and parallelize tool batch execution.
- feat: improve code block indentation normalization to handle mixed whitespace and refine markdown table rendering separators.
- feat: Add an empirical evaluation framework for measuring LLM performance and link it in the main README.
- feat: implement a new evaluation framework with test cases, metrics, and report generation, and update LLM provider integrations to support it.
- feat: Introduce coding agent settings to LLM requests to refine model behavior, implementing their application in the Anthropic provider for system prompt adjustments, prefill, message reordering, and XML document handling.
- feat: Add prefill and character reinforcement options to LLMRequest, implement Anthropic-specific handling, safety screening, and leak protection.
- feat: Add `thinking_budget` to `LLMRequest` and implement Anthropic extended thinking logic and validation.
- feat: Implement request and organization IDs for LLM responses and error metadata, enhance Anthropic error handling, and add `Refusal` finish reason.
- feat: Implement support for request-specific Anthropic beta headers and update structured output model list.
- feat: Add support for new Anthropic Claude 4 and 3.x models, enable new beta features, and refine reasoning parameter validation.
### Refactors
- refactor: Simplify tracing initialization with unwrap_or_default
- refactor: Replace map_or with is_none_or for improved clarity in MCP tool filtering
- refactor: Remove unnecessary cloning and assignment of `_updated_snapshot`.
- refactor: Inline table row rendering logic, remove duplicate parameters, and clean up the TODO list.
- refactor: update reasoning color and style for improved readability and placeholder effect
### Documentation
- docs: update changelog for v0.65.2 [skip ci]
- docs: Add a new document detailing strategies for reducing Anthropic latency and link it from the Anthropic API overview.
### Chores
- chore: update npm package.json to v0.65.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.1 version = [skip ci]
# [Version 0.65.2] - 2026-01-18
### Features
- feat: Add an empirical evaluation framework for measuring LLM performance and link it in the main README.
- feat: implement a new evaluation framework with test cases, metrics, and report generation, and update LLM provider integrations to support it.
- feat: Introduce coding agent settings to LLM requests to refine model behavior, implementing their application in the Anthropic provider for system prompt adjustments, prefill, message reordering, and XML document handling.
- feat: Add prefill and character reinforcement options to LLMRequest, implement Anthropic-specific handling, safety screening, and leak protection.
- feat: Add `thinking_budget` to `LLMRequest` and implement Anthropic extended thinking logic and validation.
- feat: Implement request and organization IDs for LLM responses and error metadata, enhance Anthropic error handling, and add `Refusal` finish reason.
- feat: Implement support for request-specific Anthropic beta headers and update structured output model list.
- feat: Add support for new Anthropic Claude 4 and 3.x models, enable new beta features, and refine reasoning parameter validation.
- feat: add tool search configuration and integration for Anthropic provider
### Refactors
- refactor: update reasoning color and style for improved readability and placeholder effect
### Documentation
- docs: Add a new document detailing strategies for reducing Anthropic latency and link it from the Anthropic API overview.
- docs: update changelog for v0.65.1 [skip ci]
- docs: update changelog for v0.65.1 [skip ci]
### Chores
- chore: update npm package.json to v0.65.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.65.0 version = [skip ci]
# [Version 0.65.1] - 2026-01-17
### Features
- feat: add tool search configuration and integration for Anthropic provider
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.65.1 [skip ci]
- docs: update changelog for v0.65.0 [skip ci]
- docs: update changelog for v0.64.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
### Chores
- chore: update npm package.json to v0.65.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update Cargo.lock
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
# [Version 0.65.1] - 2026-01-17
### Features
- feat: add tool search configuration and integration for Anthropic provider
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.65.0 [skip ci]
- docs: update changelog for v0.64.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
### Chores
- chore: update npm package.json to v0.65.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update Cargo.lock
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
# [Version 0.65.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.64.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update Cargo.lock
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.64.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore(release): bump version to {{version}}
- chore: update Cargo.lock
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.63.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update Cargo.lock
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.63.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.63.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update remaining crate versions to 0.62.0
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.63.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: bump version to 0.62.0
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.62.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.61.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.60.9 [skip ci]
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.60.9] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.61.0 [skip ci]
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.61.0] - 2026-01-17
### Features
- feat: Refactor API response handling and file search parameters, add conditional Anthropic API compilation, and enable schema generation for core types.
- feat: add Anthropic API compatibility server and documentation
### Documentation
- docs: update changelog for v0.60.8 [skip ci]
### Chores
- chore: update npm package.json to v0.60.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
# [Version 0.60.8] - 2026-01-17
### Documentation
- docs: update changelog for v0.60.7 [skip ci]
### Chores
- chore: update logo image to higher resolution
- chore: update README layout and add new logo assets
- chore: update npm package.json to v0.60.7 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: enable contributors in changelog configuration
- chore: update npm package.json to v0.60.6 version = [skip ci]
# [Version 0.60.7] - 2026-01-17
### Features
- feat: add debug logging for subagent parsing and loading; update error messages for context and segment not found
### Refactors
- refactor: remove unused LLM provider implementations
### Documentation
- docs: update changelog for v0.60.6 [skip ci]
### Chores
- chore: enable contributors in changelog configuration
- chore: update npm package.json to v0.60.6 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.60.5 version = [skip ci]
# [Version 0.60.6] - 2026-01-17
### Features
- feat: add debug logging for subagent parsing and loading; update error messages for context and segment not found
- feat: update tool policies, enhance file handling, and modify agent configuration for improved functionality
### Refactors
- refactor: remove unused LLM provider implementations
### Documentation
- docs: update changelog for v0.60.5 [skip ci]
### Chores
- chore: update npm package.json to v0.60.5 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.60.4 version = [skip ci]
# [Version 0.60.5] - 2026-01-16
### Features
- feat: update tool policies, enhance file handling, and modify agent configuration for improved functionality
- feat: update tool policies and agent configuration for improved execution control and user confirmation
- feat: enhance agent behavior configuration with Codex-inspired patterns and update tool response truncation settings
- feat: add GPT-5.2 Codex model and improve code formatting across multiple files
### Documentation
- docs: update changelog for v0.60.4 [skip ci]
### Chores
- chore: update npm package.json to v0.60.4 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.60.3 version = [skip ci]
# [Version 0.60.4] - 2026-01-16
### Features
- feat: update tool policies and agent configuration for improved execution control and user confirmation
- feat: enhance agent behavior configuration with Codex-inspired patterns and update tool response truncation settings
- feat: add GPT-5.2 Codex model and improve code formatting across multiple files
- feat: enhance sandboxing with new documentation and environment handling
### Refactors
- refactor: streamline code by simplifying conditional checks and improving output handling
- refactor: clean up code formatting and improve readability in multiple files
### Documentation
- docs: update changelog for v0.60.3 [skip ci]
### Chores
- chore: update npm package.json to v0.60.3 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.60.2 version = [skip ci]
# [Version 0.60.3] - 2026-01-14
### Features
- feat: enhance sandboxing with new documentation and environment handling
- feat: update tool policies and enhance session limit handling for tool loops
### Refactors
- refactor: streamline code by simplifying conditional checks and improving output handling
- refactor: clean up code formatting and improve readability in multiple files
- refactor: update tool policies to allow write_file and unified_file actions; remove redundant error logging
### Documentation
- docs: update changelog for v0.60.2 [skip ci]
### Chores
- chore: update npm package.json to v0.60.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update .gitignore and tool policies; change provider and API key in vtcode.toml
- chore: update npm package.json to v0.60.1 version = [skip ci]
# [Version 0.60.2] - 2026-01-10
### Features
- feat: update tool policies and enhance session limit handling for tool loops
- feat: implement planning workflow tools for managing planning workflow and enhance code block indentation normalization
- feat: implement session limit increase prompt and safety validation enhancements
### Refactors
- refactor: update tool policies to allow write_file and unified_file actions; remove redundant error logging
### Documentation
- docs: update changelog for v0.60.1 [skip ci]
### Chores
- chore: update .gitignore and tool policies; change provider and API key in vtcode.toml
- chore: update npm package.json to v0.60.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.60.0 version = [skip ci]
# [Version 0.60.1] - 2026-01-10
### Features
- feat: implement planning workflow tools for managing planning workflow and enhance code block indentation normalization
- feat: implement session limit increase prompt and safety validation enhancements
- feat: update tool policies to allow apply_patch, unified_exec, and unified_file actions
- feat: add implementation/default-agent commands for workflow control
- feat: implement Planning workflow for read-only exploration and planning
### Documentation
- docs: update changelog for v0.60.0 [skip ci]
### Chores
- chore: update npm package.json to v0.60.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update tool policies and remove unused dependencies
- chore: update npm package.json to v0.59.2 version = [skip ci]
# [Version 0.60.0] - 2026-01-10
### Features
- feat: update tool policies to allow apply_patch, unified_exec, and unified_file actions
- feat: add implementation/default-agent commands for workflow control
- feat: implement Planning workflow for read-only exploration and planning
- feat: implement sandboxing configuration and policies
- feat: enhance dynamic context discovery and update configuration
- feat: implement dynamic context discovery with file spooling for large outputs
### Documentation
- docs: update changelog for v0.59.2 [skip ci]
### Chores
- chore: update tool policies and remove unused dependencies
- chore: update npm package.json to v0.59.2 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.59.1 version = [skip ci]
# [Version 0.59.2] - 2026-01-08
### Features
- feat: implement sandboxing configuration and policies
- feat: enhance dynamic context discovery and update configuration
- feat: implement dynamic context discovery with file spooling for large outputs
### Refactors
- refactor: update tool policies, enhance subagent cleanup, and improve documentation
### Documentation
- docs: update changelog for v0.59.1 [skip ci]
### Chores
- chore: update npm package.json to v0.59.1 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.59.0 version = [skip ci]
# [Version 0.59.1] - 2026-01-07
### Refactors
- refactor: update tool policies, enhance subagent cleanup, and improve documentation
- refactor: clean up code formatting and improve readability across multiple files
### Documentation
- docs: update changelog for v0.59.0 [skip ci]
- docs: update changelog for v0.58.26 [skip ci]
- docs: update ACP V2 Migration Guide for improved clarity and formatting
### Chores
- chore: update npm package.json to v0.59.0 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.25 version = [skip ci]
# [Version 0.59.0] - 2026-01-06
### Features
- feat: add LRU cache for canonicalized paths and optimize vector allocations
- feat: restore Kitty keyboard protocol support and update session handling
### Refactors
- refactor: clean up code formatting and improve readability across multiple files
### Documentation
- docs: update changelog for v0.58.26 [skip ci]
- docs: update ACP V2 Migration Guide for improved clarity and formatting
- docs: update changelog for v0.58.25 [skip ci]
### Chores
- chore: update npm package.json to v0.58.25 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.24 version = [skip ci]
# [Version 0.58.26] - 2026-01-06
### Features
- feat: add LRU cache for canonicalized paths and optimize vector allocations
- feat: restore Kitty keyboard protocol support and update session handling
### Refactors
- refactor: clean up code formatting and improve readability across multiple files
### Documentation
- docs: update ACP V2 Migration Guide for improved clarity and formatting
- docs: update changelog for v0.58.25 [skip ci]
### Chores
- chore: update npm package.json to v0.58.25 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.24 version = [skip ci]
# [Version 0.58.25] - 2026-01-06
### Features
- feat: add LRU cache for canonicalized paths and optimize vector allocations
- feat: restore Kitty keyboard protocol support and update session handling
### Documentation
- docs: update changelog for v0.58.24 [skip ci]
### Chores
- chore: update npm package.json to v0.58.24 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.23 version = [skip ci]
# [Version 0.58.24] - 2026-01-05
### Documentation
- docs: update changelog for v0.58.23 [skip ci]
### Chores
- chore: update npm package.json to v0.58.23 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.22 version = [skip ci]
# [Version 0.58.23] - 2026-01-05
### Documentation
- docs: update changelog for v0.58.22 [skip ci]
### Chores
- chore: update npm package.json to v0.58.22 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.21 version = [skip ci]
# [Version 0.58.22] - 2026-01-04
### Documentation
- docs: update changelog for v0.58.21 [skip ci]
### Chores
- chore: update npm package.json to v0.58.21 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.20 version = [skip ci]
# [Version 0.58.21] - 2026-01-04
### Bug Fixes
- fix: suppress dead code warnings for unused UI and agent functions
### Documentation
- docs: update changelog for v0.58.20 [skip ci]
### Chores
- chore: update npm package.json to v0.58.20 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.19 version = [skip ci]
# [Version 0.58.20] - 2026-01-04
### Bug Fixes
- fix: suppress dead code warnings for unused UI and agent functions
- fix: prefix unused variable with underscore in config_watcher
### Documentation
- docs: update changelog for v0.58.19 [skip ci]
### Chores
- chore: update npm package.json to v0.58.19 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.18 version = [skip ci]
# [Version 0.58.19] - 2026-01-04
### Features
- feat(core): enhance tool caching and UI redraw optimization
### Bug Fixes
- fix: prefix unused variable with underscore in config_watcher
- fix: remove dead code and fix compilation errors
### Documentation
- docs: update changelog for v0.58.18 [skip ci]
- docs: update changelog for v0.58.17 [skip ci]
- docs: update changelog for v0.58.16 [skip ci]
### Chores
- chore: update npm package.json to v0.58.18 version = [skip ci]
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.15 version = [skip ci]
# [Version 0.58.18] - 2026-01-04
### Features
- feat(core): enhance tool caching and UI redraw optimization
- feat(core): add file system watcher and performance optimization infrastructure
- feat(core): integrate real performance optimizations into tool registry
### Bug Fixes
- fix: remove dead code and fix compilation errors
### Documentation
- docs: update changelog for v0.58.17 [skip ci]
- docs: update changelog for v0.58.16 [skip ci]
- docs: update changelog for v0.58.15 [skip ci]
- docs: reorganize documentation and integrate skill tools into registry
### Chores
- chore(release): bump version to {{version}}
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.15 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.14 version = [skip ci]
# [Version 0.58.17] - 2026-01-04
### Features
- feat(core): enhance tool caching and UI redraw optimization
- feat(core): add file system watcher and performance optimization infrastructure
- feat(core): integrate real performance optimizations into tool registry
### Documentation
- docs: update changelog for v0.58.16 [skip ci]
- docs: update changelog for v0.58.15 [skip ci]
- docs: reorganize documentation and integrate skill tools into registry
### Chores
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.15 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.14 version = [skip ci]
# [Version 0.58.16] - 2026-01-04
### Features
- feat(core): enhance tool caching and UI redraw optimization
- feat(core): add file system watcher and performance optimization infrastructure
- feat(core): integrate real performance optimizations into tool registry
### Documentation
- docs: update changelog for v0.58.15 [skip ci]
- docs: reorganize documentation and integrate skill tools into registry
### Chores
- chore: update npm package.json to v0.58.15 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.14 version = [skip ci]
# [Version 0.58.15] - 2026-01-04
### Features
- feat(core): add file system watcher and performance optimization infrastructure
- feat(core): integrate real performance optimizations into tool registry
### Documentation
- docs: reorganize documentation and integrate skill tools into registry
- docs: update changelog for v0.58.14 [skip ci]
### Chores
- chore: update npm package.json to v0.58.14 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: remove unused GeminiPromptCacheMode import.
- chore: update npm package.json to v0.58.13 version = [skip ci]
# [Version 0.58.14] - 2026-01-03
### Features
- feat: enhance LLM provider initialization with client injection and refine prompt caching
- feat: Refactor tool registry to use MCP tool index cache and update mutability of inventory and tool policy access.
- feat: Implement parallel tool execution for agent actions and update tool registry operations to be asynchronous.
- feat: Refactor tool permission context and enhance command safety validation with new progress updates for tool execution.
### Refactors
- refactor: simplify nested conditional logic with chained `&& let` patterns
- refactor: update symbol name extraction to use `ChildByField` and add Rust language test.
- refactor: Extract progress update guard and elapsed time updater to `progress.rs` and add `PlaceholderSpinner::force_refresh`.
### Documentation
- docs: update changelog for v0.58.13 [skip ci]
### Style Changes
- style: apply consistent formatting and whitespace adjustments
### Chores
- chore: remove unused GeminiPromptCacheMode import.
- chore: update npm package.json to v0.58.13 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.12 version = [skip ci]
# [Version 0.58.13] - 2026-01-03
### Features
- feat: enhance LLM provider initialization with client injection and refine prompt caching
- feat: Refactor tool registry to use MCP tool index cache and update mutability of inventory and tool policy access.
- feat: Implement parallel tool execution for agent actions and update tool registry operations to be asynchronous.
- feat: Refactor tool permission context and enhance command safety validation with new progress updates for tool execution.
### Bug Fixes
- fix: resolve Windows build errors in vtcode-core
### Refactors
- refactor: simplify nested conditional logic with chained `&& let` patterns
- refactor: update symbol name extraction to use `ChildByField` and add Rust language test.
- refactor: Extract progress update guard and elapsed time updater to `progress.rs` and add `PlaceholderSpinner::force_refresh`.
### Documentation
- docs: update changelog for v0.58.12 [skip ci]
- docs: add Windows build fixes documentation
### Style Changes
- style: apply consistent formatting and whitespace adjustments
### Chores
- chore: update npm package.json to v0.58.12 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.11 version = [skip ci]
# [Version 0.58.12] - 2026-01-02
### Bug Fixes
- fix: resolve Windows build errors in vtcode-core
- fix: suppress dead_code warnings for planned/stub functions
### Documentation
- docs: add Windows build fixes documentation
- docs: update changelog for v0.58.11 [skip ci]
### Chores
- chore: update npm package.json to v0.58.11 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.10 version = [skip ci]
# [Version 0.58.11] - 2026-01-02
### Bug Fixes
- fix: suppress dead_code warnings for planned/stub functions
- fix: build-release workflow now triggers on tag push events
### Documentation
- docs: update changelog for v0.58.10 [skip ci]
### Chores
- chore: update npm package.json to v0.58.10 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.9 version = [skip ci]
# [Version 0.58.10] - 2026-01-02
### Bug Fixes
- fix: build-release workflow now triggers on tag push events
### Documentation
- docs: update changelog for v0.58.9 [skip ci]
### Chores
- chore: update npm package.json to v0.58.9 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.8 version = [skip ci]
# [Version 0.58.9] - 2026-01-02
### Features
- feat: auto-trigger build-release workflow on GitHub release creation
### Documentation
- docs: update changelog for v0.58.8 [skip ci]
### Chores
- chore: update npm package.json to v0.58.8 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.7 version = [skip ci]
# [Version 0.58.8] - 2026-01-02
### Features
- feat: auto-trigger build-release workflow on GitHub release creation
### Bug Fixes
- fix: suppress unused_imports warning in openai.rs for CI compatibility
- fix: install OpenSSL dependencies for Linux builds in CI
- fix: conditionally import debug-only items to fix release build
### Documentation
- docs: update changelog for v0.58.7 [skip ci]
- docs: add instructions for manually triggering release build
- docs: add quick reference for monitoring and auto-install
- docs: add release monitoring guide with auto-install instructions
- docs: add native installer readme - central documentation hub
- docs: deployment complete - v0.58.6 release ready
- docs: add release v0.58.6 and installer test guide
### Chores
- chore: update npm package.json to v0.58.7 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.6 version = [skip ci]
# [Version 0.58.7] - 2026-01-02
### Bug Fixes
- fix: suppress unused_imports warning in openai.rs for CI compatibility
- fix: install OpenSSL dependencies for Linux builds in CI
- fix: conditionally import debug-only items to fix release build
- fix: ensure get_download_url outputs only URL to stdout
- fix: redirect all logging to stderr in installer script
### Documentation
- docs: add instructions for manually triggering release build
- docs: add quick reference for monitoring and auto-install
- docs: add release monitoring guide with auto-install instructions
- docs: add native installer readme - central documentation hub
- docs: deployment complete - v0.58.6 release ready
- docs: add release v0.58.6 and installer test guide
- docs: update changelog for v0.58.6 [skip ci]
- docs: add native installer implementation status report
### Chores
- chore: update npm package.json to v0.58.6 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: update npm package.json to v0.58.5 version = [skip ci]
# [Version 0.58.6] - 2026-01-02
### Features
- feat: add native installer with auto-updater module
### Bug Fixes
- fix: ensure get_download_url outputs only URL to stdout
- fix: redirect all logging to stderr in installer script
- fix: correct Python variable substitution in Homebrew workflow
- fix: improve release.toml commit message template for consistency
### Documentation
- docs: add native installer implementation status report
- docs: update changelog for v0.58.5 [skip ci]
- docs: add .nojekyll to bypass Jekyll processing
- docs: remove HTML index, use Jekyll markdown
- docs: add HTML landing page for GitHub Pages
- docs: add Jekyll config and documentation index
- docs: update Homebrew documentation - simplified architecture
- docs: add Homebrew verification checklist - release automation complete
- docs: add actionable next steps for completing homebrew distribution setup
- docs: add comprehensive homebrew fix summary with all solutions applied
- docs: explain why homebrew updates stopped and root cause analysis
- docs: add guide for setting up custom homebrew tap repository
### Chores
- chore: update npm package.json to v0.58.5 version = [skip ci]
- chore(release): bump version to {{version}}
- chore: remove Pages workflow - using simpler direct branch deployment
- chore: remove redundant release-on-tag.yml workflow
- chore: update npm package.json to v0.58.4 version = [skip ci]
# [Version 0.58.5] - 2026-01-02
### Features
- feat: add native installer with auto-updater module
### Bug Fixes
- fix: correct Python variable substitution in Homebrew workflow
- fix: improve release.toml commit message template for consistency
- fix: resolve python string interpolation in homebrew formula updates
- fix: improve homebrew formula regex patterns for reliable checksum updates
- fix: homebrew release automation and YAML workflow indentation
### Documentation
- docs: add .nojekyll to bypass Jekyll processing
- docs: remove HTML index, use Jekyll markdown
- docs: add HTML landing page for GitHub Pages
- docs: add Jekyll config and documentation index
- docs: update Homebrew documentation - simplified architecture
- docs: add Homebrew verification checklist - release automation complete
- docs: add actionable next steps for completing homebrew distribution setup
- docs: add comprehensive homebrew fix summary with all solutions applied
- docs: explain why homebrew updates stopped and root cause analysis
- docs: add guide for setting up custom homebrew tap repository
- docs: update changelog for v0.58.4 [skip ci]
- docs: add verification summary for homebrew release fixes
### Chores
- chore: remove Pages workflow - using simpler direct branch deployment
- chore: remove redundant release-on-tag.yml workflow
- chore: update npm package.json to v0.58.4 version = [skip ci]
- chore: release v{{version}}
- chore: update npm package.json to v0.58.3 version = [skip ci]
# [Version 0.58.4] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
### Bug Fixes
- fix: resolve python string interpolation in homebrew formula updates
- fix: improve homebrew formula regex patterns for reliable checksum updates
- fix: homebrew release automation and YAML workflow indentation
- fix: update dependencies and improve error handling in terminal functions
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
### Documentation
- docs: add verification summary for homebrew release fixes
- docs: update changelog for v0.58.3 [skip ci]
- docs: update changelog for v0.58.2 [skip ci]
- docs: update changelog for v0.58.1 [skip ci]
- docs: update changelog for v0.58.0 [skip ci]
- docs: update changelog for v0.57.0 [skip ci]
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
### Chores
- chore: update npm package.json to v0.58.3 version = [skip ci]
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore(deps): bump the all-rust-deps group with 14 updates
# [Version 0.58.3] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
### Bug Fixes
- fix: update dependencies and improve error handling in terminal functions
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
### Documentation
- docs: update changelog for v0.58.2 [skip ci]
- docs: update changelog for v0.58.1 [skip ci]
- docs: update changelog for v0.58.0 [skip ci]
- docs: update changelog for v0.57.0 [skip ci]
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
### Chores
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore(deps): bump the all-rust-deps group with 14 updates
- chore: update npm package.json to v0.55.1 [skip ci]
# [Version 0.58.2] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
### Documentation
- docs: update changelog for v0.58.1 [skip ci]
- docs: update changelog for v0.58.0 [skip ci]
- docs: update changelog for v0.57.0 [skip ci]
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
### Chores
- chore: release v{{version}}
- chore: release v{{version}}
- chore: release v{{version}}
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore(deps): bump the all-rust-deps group with 14 updates
- chore: update npm package.json to v0.55.1 [skip ci]
# [Version 0.58.1] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.58.0 [skip ci]
- docs: update changelog for v0.57.0 [skip ci]
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
- docs: update changelog for v0.55.1 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
### Chores
- chore: release v{{version}}
- chore: release v{{version}}
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore: update npm package.json to v0.55.1 [skip ci]
- chore: release v0.55.1
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
# [Version 0.58.0] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.57.0 [skip ci]
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
- docs: update changelog for v0.55.1 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
### Chores
- chore: release v{{version}}
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore: update npm package.json to v0.55.1 [skip ci]
- chore: release v0.55.1
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
# [Version 0.57.0] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
- docs: update changelog for v0.55.1 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
### Chores
- chore: fix release metadata for vtcode-file-search and vtcode-process-hardening, add version constraint
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore: update npm package.json to v0.55.1 [skip ci]
- chore: release v0.55.1
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
# [Version 0.56.0] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.56.0 [skip ci]
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
- docs: update changelog for v0.55.1 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
### Chores
- chore: add missing metadata to vtcode-file-search and vtcode-process-hardening
- chore: update npm package.json to v0.55.1 [skip ci]
- chore: release v0.55.1
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
# [Version 0.56.0] - 2026-01-02
### Features
- feat(runner): add keyboard protocol environment check and enhance logging
- feat(tools): Add move and copy file operations with error handling
- feat(command-safety): Enhance command safety module with comprehensive integration and documentation
- feat(prompts): Revise system prompts for clarity, autonomy, and Codex alignment
- feat(dependencies): Update Cargo.lock and Cargo.toml with new packages and versions
- feat(ollama-integration): Implement comprehensive Ollama client and progress reporting system
- feat(process-hardening): Introduce vtcode-process-hardening crate for enhanced security measures
- feat(file-search): Implement vtcode-file-search crate and integration with extensions
- feat: Unify system prompt instruction generation and skill rendering via `get_user_instructions` and `AgentConfig`.
- feat: Improve context manager performance with incremental stats and bolster tool security with path and command validation.
- feat: Implement the Desire Paths philosophy by updating agent prompts and documentation to improve agent UX.
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
### Bug Fixes
- fix(models): Correct ClaudeOpus41 version and update related assertions refactor(loader): Adjust config loading order and clarify comments feat(output-styles): Add kebab-case renaming for OutputStyleFileConfig
### Refactors
- refactor(config): Remove ConfigOptimizer and implement ConfigBuilder for streamlined configuration management
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: Add Phase 3 extension integration planning and file search documentation
- docs: Update implementation summary and configuration for file search and system prompt enhancements
- docs: update changelog for v0.55.1 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
### Chores
- chore: update npm package.json to v0.55.1 [skip ci]
- chore: release v0.55.1
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
# [Version 0.55.1] - 2025-12-29
### Features
- feat: add tokio dependency and improve error handling in background task execution
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
- feat(notifications): add toggle for terminal notifications in config
- feat(cli): add support for multiple workspaces and enhanced security controls
- feat(release): enhance GitHub account handling for CI environments
### Refactors
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.55.0 [skip ci]
- docs: update changelog for v0.54.4 [skip ci]
- docs: update changelog for v0.54.3 [skip ci]
### Chores
- chore: release v0.55.0
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
- chore: release v0.54.3
- chore: update npm package.json to v0.54.2 [skip ci]
# [Version 0.55.0] - 2025-12-29
### Features
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
- feat(notifications): add toggle for terminal notifications in config
- feat(cli): add support for multiple workspaces and enhanced security controls
- feat(release): enhance GitHub account handling for CI environments
### Refactors
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.54.4 [skip ci]
- docs: update changelog for v0.54.3 [skip ci]
### Chores
- chore: release v0.54.4
- chore: update npm package.json to v0.54.3 [skip ci]
- chore: release v0.54.3
- chore: update npm package.json to v0.54.2 [skip ci]
# [Version 0.54.4] - 2025-12-29
### Features
- feat(subagents): document subagent system and built-in agents; update README and changelog
- feat(statusline): add custom status line scripts and JSON input handling
- feat: add reverse search and background operation handling in TUI
- feat(output-styles): implement output styles feature with customizable behavior and response formats
- feat(hooks): add lifecycle hooks for file protection, command logging, code formatting, markdown formatting, and notifications
- feat(marketplace): add marketplace and plugin management commands
- feat(notifications): add toggle for terminal notifications in config
- feat(cli): add support for multiple workspaces and enhanced security controls
- feat(release): enhance GitHub account handling for CI environments
### Refactors
- refactor: improve Linux checksum handling and release verification in scripts
### Documentation
- docs: update changelog for v0.54.3 [skip ci]
### Chores
- chore: update npm package.json to v0.54.3 [skip ci]
- chore: release v0.54.3
- chore: update npm package.json to v0.54.2 [skip ci]
# [Version 0.54.3] - 2025-12-28
### Features
- feat(subagents): add subagent system for delegating tasks to specialized agents
- Built-in subagents: explore (haiku, read-only), plan (sonnet, research), general (sonnet, full), code-reviewer, debugger
- `spawn_subagent` tool with resume, thoroughness, parent_context params
- Custom agents via Markdown with YAML frontmatter in `.vtcode/agents/` or `~/.vtcode/agents/`
- System prompts updated to guide orchestrator delegation
- Documentation: `docs/subagents/SUBAGENTS.md`
- feat(notifications): add toggle for terminal notifications in config
- feat(cli): add support for multiple workspaces and enhanced security controls
- feat(release): enhance GitHub account handling for CI environments
- feat(a2a): complete CLI integration and documentation\n\n- Add full A2A CLI with serve, discover, send-task, list-tasks, get-task, cancel-task commands\n- Create comprehensive CLI handlers for all A2A operations\n- Fix streaming event handling with proper pinning\n- Update server.rs Box<dyn Stream> return type for axum compatibility\n- Add completion summary document\n- All checks pass: cargo check --package vtcode-core\n\nImplements: A2A Protocol Phase 4 - CLI integration and user-facing features
- feat(a2a): add A2A client with streaming support\n\n- New A2aClient for discovery, task ops, push config, and streaming\n- SSE client parses streaming events without extra deps\n- Simple incremental request IDs and HTTPS agent card fetch\n- Tests added for SSE parsing helpers\n\nTests: cargo test --package vtcode-core --lib a2a (39/39)
- feat(a2a): trigger webhooks on streaming events\n\n- Add webhook_notifier to server state and wire into streaming pipeline\n- Fire webhooks for status updates and messages when broadcasted\n- Fix SSRF-safe config retrieval and avoid Option to_string() error\n- Clean up unused tracing import in webhook module\n- Tests: all A2A suites pass (37/37)
- feat(a2a): finish push notification config storage and RPC wiring\n\n- Add webhook config storage to TaskManager (set/get/remove) with SSRF validation\n- Wire JSON-RPC handlers for pushNotificationConfig set/get\n- Fix server dispatch and imports\n- All A2A tests pass (37/37) including server + webhook
- feat(a2a): add webhook notifier for push notifications (Phase 3.2 partial)
- feat(a2a): implement full SSE streaming support (Phase 3.1)
- feat: implement Agent2Agent (A2A) Protocol support (Phase 1 & 2)
- feat: Add async method to InlineSession for receiving next event
- feat: Remove the `plan` tool and associated components, and update related tool and skill management logic.
- feat: Refactor tool permission handling for TUI-only execution and update default LLM provider configuration.
- feat: Add GitHub account switching and cleanup functionality in release script
### Refactors
- refactor(a2a): clean up unused imports and improve webhook handling
- refactor: Enhance analysis command to support multiple analysis types and improve error handling
- refactor: Update tool policies to prompt-based for MCP time functions and improve session handling with cancellation support
- refactor: Remove unused agent diagnostic tools from TODO documentation
- refactor: Refine tool policies by removing unused tools, changing several to prompt-based, and making `wrap_text` test-only.
- refactor: overhaul TUI, tool policy, and context management, adding new documentation and tests.
- refactor: Remove token budget management and related token estimation/truncation components, and add associated documentation and verification scripts.
- refactor: improve error message for missing MCP tools with installation instructions
### Documentation
- docs: update changelog for v0.54.2 [skip ci]
- docs: update changelog for v0.55.0 [skip ci]
- docs(a2a): add comprehensive documentation for A2A Protocol implementation
- docs(a2a): add Phase 3 implementation status tracker
### Chores
- chore: update npm package.json to v0.54.2 [skip ci]
- chore: release v0.54.2
- chore: update npm package.json to v0.54.1 [skip ci]
# [Version 0.54.2] - 2025-12-28
### Features
- feat(a2a): complete CLI integration and documentation\n\n- Add full A2A CLI with serve, discover, send-task, list-tasks, get-task, cancel-task commands\n- Create comprehensive CLI handlers for all A2A operations\n- Fix streaming event handling with proper pinning\n- Update server.rs Box<dyn Stream> return type for axum compatibility\n- Add completion summary document\n- All checks pass: cargo check --package vtcode-core\n\nImplements: A2A Protocol Phase 4 - CLI integration and user-facing features
- feat(a2a): add A2A client with streaming support\n\n- New A2aClient for discovery, task ops, push config, and streaming\n- SSE client parses streaming events without extra deps\n- Simple incremental request IDs and HTTPS agent card fetch\n- Tests added for SSE parsing helpers\n\nTests: cargo test --package vtcode-core --lib a2a (39/39)
- feat(a2a): trigger webhooks on streaming events\n\n- Add webhook_notifier to server state and wire into streaming pipeline\n- Fire webhooks for status updates and messages when broadcasted\n- Fix SSRF-safe config retrieval and avoid Option to_string() error\n- Clean up unused tracing import in webhook module\n- Tests: all A2A suites pass (37/37)
- feat(a2a): finish push notification config storage and RPC wiring\n\n- Add webhook config storage to TaskManager (set/get/remove) with SSRF validation\n- Wire JSON-RPC handlers for pushNotificationConfig set/get\n- Fix server dispatch and imports\n- All A2A tests pass (37/37) including server + webhook
- feat(a2a): add webhook notifier for push notifications (Phase 3.2 partial)
- feat(a2a): implement full SSE streaming support (Phase 3.1)
- feat: implement Agent2Agent (A2A) Protocol support (Phase 1 & 2)
- feat: Add async method to InlineSession for receiving next event
- feat: Remove the `plan` tool and associated components, and update related tool and skill management logic.
- feat: Refactor tool permission handling for TUI-only execution and update default LLM provider configuration.
- feat: Add GitHub account switching and cleanup functionality in release script
- feat: Introduce `EnhancedSkillLoader` and `EnhancedSkill` for unified skill and tool management, and refactor skill discovery results across the agent and CLI.
- feat: add support for loading skill `references/` and `assets/` directories and introduce `ResourceType::Asset`
- feat: Reimplement skill management with a new skill model and dedicated modules.
- feat: Refine tool policies and skill loading for lazy-loaded capabilities, updating system prompts to reflect on-demand activation.
- feat: Implement lazy-loading and tiered disclosure for agent skills and tools, reducing default available tools and updating system prompts.
- feat: Introduce skill varieties and enhance skill listing/loading with filtering and dormant tool support
- feat: Implement `Tool` trait for `CliToolBridge` and integrate skill-based tool registration with `ToolRegistry`.
- feat: Add new Ollama cloud models and update reasoning model detection.
- feat: Implement on-demand skill loading with `LoadSkillTool` and `LoadSkillResourceTool`, and enable skill restoration from previous sessions.
- feat: Add `ListSkillsTool` for programmatic skill discovery, replacing direct skill prompt integration.
- feat: Add `LoadSkillTool` for progressive skill instruction loading, enhance skill context with path storage, and integrate skill discovery into agent setup.
- feat: introduce context summarization with adaptive trimming integration and new `Summarize` retention choice.
- feat: calculate context usage from history and add a final pre-request safety check after trimming.
- feat: Add `mcp::fetch` and `mcp::time` tools, simplify LSP client message handling, and remove outdated agent system analysis from TODO documentation.
- feat: Implement PTY session termination on Ctrl+C cancellation with debounced signal handling and status line feedback.
- feat: Implement timed double Ctrl+C for agent exit, deferring shutdown, and update the default model.
- feat: Add `--skip-release` option and enhance GitHub release verification logic with CI environment detection.
### Refactors
- refactor(a2a): clean up unused imports and improve webhook handling
- refactor: Enhance analysis command to support multiple analysis types and improve error handling
- refactor: Update tool policies to prompt-based for MCP time functions and improve session handling with cancellation support
- refactor: Remove unused agent diagnostic tools from TODO documentation
- refactor: Refine tool policies by removing unused tools, changing several to prompt-based, and making `wrap_text` test-only.
- refactor: overhaul TUI, tool policy, and context management, adding new documentation and tests.
- refactor: Remove token budget management and related token estimation/truncation components, and add associated documentation and verification scripts.
- refactor: improve error message for missing MCP tools with installation instructions
- refactor: Introduce a dedicated interaction loop for centralized user input and turn flow, updating session and context management.
- refactor: consistently use `adaptive_trim` with `pruning_ledger` across all proactive token budget guards.
- refactor: Restructure agent turn execution with new guard, context, and tool outcome modules, removing old loop detection, and updating LSP tools.
### Documentation
- docs: update changelog for v0.55.0 [skip ci]
- docs(a2a): add comprehensive documentation for A2A Protocol implementation
- docs(a2a): add Phase 3 implementation status tracker
- docs: update changelog for v0.54.1 [skip ci]
- docs: Streamline TODO by removing verbose system skill enumeration and adding a focused task.
### Chores
- chore: update npm package.json to v0.54.1 [skip ci]
- chore: release v0.54.1
- chore: update npm package.json to v0.54.0 [skip ci]
# [Version 0.55.0] - 2025-12-28
### Features
- feat(a2a): complete CLI integration and documentation\n\n- Add full A2A CLI with serve, discover, send-task, list-tasks, get-task, cancel-task commands\n- Create comprehensive CLI handlers for all A2A operations\n- Fix streaming event handling with proper pinning\n- Update server.rs Box<dyn Stream> return type for axum compatibility\n- Add completion summary document\n- All checks pass: cargo check --package vtcode-core\n\nImplements: A2A Protocol Phase 4 - CLI integration and user-facing features
- feat(a2a): add A2A client with streaming support\n\n- New A2aClient for discovery, task ops, push config, and streaming\n- SSE client parses streaming events without extra deps\n- Simple incremental request IDs and HTTPS agent card fetch\n- Tests added for SSE parsing helpers\n\nTests: cargo test --package vtcode-core --lib a2a (39/39)
- feat(a2a): trigger webhooks on streaming events\n\n- Add webhook_notifier to server state and wire into streaming pipeline\n- Fire webhooks for status updates and messages when broadcasted\n- Fix SSRF-safe config retrieval and avoid Option to_string() error\n- Clean up unused tracing import in webhook module\n- Tests: all A2A suites pass (37/37)
- feat(a2a): finish push notification config storage and RPC wiring\n\n- Add webhook config storage to TaskManager (set/get/remove) with SSRF validation\n- Wire JSON-RPC handlers for pushNotificationConfig set/get\n- Fix server dispatch and imports\n- All A2A tests pass (37/37) including server + webhook
- feat(a2a): add webhook notifier for push notifications (Phase 3.2 partial)
- feat(a2a): implement full SSE streaming support (Phase 3.1)
- feat: implement Agent2Agent (A2A) Protocol support (Phase 1 & 2)
- feat: Add async method to InlineSession for receiving next event
- feat: Remove the `plan` tool and associated components, and update related tool and skill management logic.
- feat: Refactor tool permission handling for TUI-only execution and update default LLM provider configuration.
- feat: Add GitHub account switching and cleanup functionality in release script
- feat: Introduce `EnhancedSkillLoader` and `EnhancedSkill` for unified skill and tool management, and refactor skill discovery results across the agent and CLI.
- feat: add support for loading skill `references/` and `assets/` directories and introduce `ResourceType::Asset`
- feat: Reimplement skill management with a new skill model and dedicated modules.
- feat: Refine tool policies and skill loading for lazy-loaded capabilities, updating system prompts to reflect on-demand activation.
- feat: Implement lazy-loading and tiered disclosure for agent skills and tools, reducing default available tools and updating system prompts.
- feat: Introduce skill varieties and enhance skill listing/loading with filtering and dormant tool support
- feat: Implement `Tool` trait for `CliToolBridge` and integrate skill-based tool registration with `ToolRegistry`.
- feat: Add new Ollama cloud models and update reasoning model detection.
- feat: Implement on-demand skill loading with `LoadSkillTool` and `LoadSkillResourceTool`, and enable skill restoration from previous sessions.
- feat: Add `ListSkillsTool` for programmatic skill discovery, replacing direct skill prompt integration.
- feat: Add `LoadSkillTool` for progressive skill instruction loading, enhance skill context with path storage, and integrate skill discovery into agent setup.
- feat: introduce context summarization with adaptive trimming integration and new `Summarize` retention choice.
- feat: calculate context usage from history and add a final pre-request safety check after trimming.
- feat: Add `mcp::fetch` and `mcp::time` tools, simplify LSP client message handling, and remove outdated agent system analysis from TODO documentation.
- feat: Implement PTY session termination on Ctrl+C cancellation with debounced signal handling and status line feedback.
- feat: Implement timed double Ctrl+C for agent exit, deferring shutdown, and update the default model.
- feat: Add `--skip-release` option and enhance GitHub release verification logic with CI environment detection.
### Refactors
- refactor(a2a): clean up unused imports and improve webhook handling
- refactor: Enhance analysis command to support multiple analysis types and improve error handling
- refactor: Update tool policies to prompt-based for MCP time functions and improve session handling with cancellation support
- refactor: Remove unused agent diagnostic tools from TODO documentation
- refactor: Refine tool policies by removing unused tools, changing several to prompt-based, and making `wrap_text` test-only.
- refactor: overhaul TUI, tool policy, and context management, adding new documentation and tests.
- refactor: Remove token budget management and related token estimation/truncation components, and add associated documentation and verification scripts.
- refactor: improve error message for missing MCP tools with installation instructions
- refactor: Introduce a dedicated interaction loop for centralized user input and turn flow, updating session and context management.
- refactor: consistently use `adaptive_trim` with `pruning_ledger` across all proactive token budget guards.
- refactor: Restructure agent turn execution with new guard, context, and tool outcome modules, removing old loop detection, and updating LSP tools.
### Documentation
- docs(a2a): add comprehensive documentation for A2A Protocol implementation
- docs(a2a): add Phase 3 implementation status tracker
- docs: update changelog for v0.54.1 [skip ci]
- docs: Streamline TODO by removing verbose system skill enumeration and adding a focused task.
### Chores
- chore: update npm package.json to v0.54.1 [skip ci]
- chore: release v0.54.1
- chore: update npm package.json to v0.54.0 [skip ci]
# [Version 0.54.1] - 2025-12-27
### Features
- feat: Introduce `EnhancedSkillLoader` and `EnhancedSkill` for unified skill and tool management, and refactor skill discovery results across the agent and CLI.
- feat: add support for loading skill `references/` and `assets/` directories and introduce `ResourceType::Asset`
- feat: Reimplement skill management with a new skill model and dedicated modules.
- feat: Refine tool policies and skill loading for lazy-loaded capabilities, updating system prompts to reflect on-demand activation.
- feat: Implement lazy-loading and tiered disclosure for agent skills and tools, reducing default available tools and updating system prompts.
- feat: Introduce skill varieties and enhance skill listing/loading with filtering and dormant tool support
- feat: Implement `Tool` trait for `CliToolBridge` and integrate skill-based tool registration with `ToolRegistry`.
- feat: Add new Ollama cloud models and update reasoning model detection.
- feat: Implement on-demand skill loading with `LoadSkillTool` and `LoadSkillResourceTool`, and enable skill restoration from previous sessions.
- feat: Add `ListSkillsTool` for programmatic skill discovery, replacing direct skill prompt integration.
- feat: Add `LoadSkillTool` for progressive skill instruction loading, enhance skill context with path storage, and integrate skill discovery into agent setup.
- feat: introduce context summarization with adaptive trimming integration and new `Summarize` retention choice.
- feat: calculate context usage from history and add a final pre-request safety check after trimming.
- feat: Add `mcp::fetch` and `mcp::time` tools, simplify LSP client message handling, and remove outdated agent system analysis from TODO documentation.
- feat: Implement PTY session termination on Ctrl+C cancellation with debounced signal handling and status line feedback.
- feat: Implement timed double Ctrl+C for agent exit, deferring shutdown, and update the default model.
- feat: Add `--skip-release` option and enhance GitHub release verification logic with CI environment detection.
- feat: Implement LSP client and manager with agent slash commands, and add LLM provider caching tests.
### Refactors
- refactor: Introduce a dedicated interaction loop for centralized user input and turn flow, updating session and context management.
- refactor: consistently use `adaptive_trim` with `pruning_ledger` across all proactive token budget guards.
- refactor: Restructure agent turn execution with new guard, context, and tool outcome modules, removing old loop detection, and updating LSP tools.
### Documentation
- docs: Streamline TODO by removing verbose system skill enumeration and adding a focused task.
- docs: update changelog for v0.54.0 [skip ci]
### Chores
- chore: update npm package.json to v0.54.0 [skip ci]
- chore: release v0.54.0
- chore: update npm package.json to v0.53.2 [skip ci]
# [Version 0.54.0] - 2025-12-27
### Features
- feat: Implement LSP client and manager with agent slash commands, and add LLM provider caching tests.
- feat: Update default agent configuration to HuggingFace and refine tool schemas and prompt generation logic.
- feat: Introduce dynamic system prompt enhancements including temporal context and working directory awareness, along with refined tool usage guidelines for improved agent performance.
- feat: Enhance textual tool call parsing, pre-validate arguments, and refine tool failure detection to improve agent robustness.
### Documentation
- docs: update changelog for v0.53.2 [skip ci]
### Chores
- chore: update npm package.json to v0.53.2 [skip ci]
- chore: release v0.53.2
- chore: update npm package.json to v0.53.1 [skip ci]
# [Version 0.53.2] - 2025-12-26
### Features
- feat: Update default agent configuration to HuggingFace and refine tool schemas and prompt generation logic.
- feat: Introduce dynamic system prompt enhancements including temporal context and working directory awareness, along with refined tool usage guidelines for improved agent performance.
- feat: Enhance textual tool call parsing, pre-validate arguments, and refine tool failure detection to improve agent robustness.
- feat: Enhance session resume/fork logic and improve conversation history display during session startup.
- feat: add session resumption functionality and update related actions
- feat: implement session forking with custom session ID support
- feat: enhance documentation and prompts for clarity, consistency, and performance improvements
- feat: optimize ANSI syntax highlighting in diff renderer for improved performance
### Refactors
- refactor: update reasoning labels for clarity in justification and session headers
- refactor: streamline toolset by merging agent diagnostics and removing deprecated tools
- refactor: simplify error handling in dotenv loading
### Documentation
- docs: update changelog for v0.53.1 [skip ci]
### Chores
- chore: update npm package.json to v0.53.1 [skip ci]
- chore: release v0.53.1
- chore: remove completed tasks from TODO.md and improve memory usage for large conversations
- chore: update npm package.json to v0.53.0 [skip ci]
# [Version 0.53.1] - 2025-12-26
### Features
- feat: Enhance session resume/fork logic and improve conversation history display during session startup.
- feat: add session resumption functionality and update related actions
- feat: implement session forking with custom session ID support
- feat: enhance documentation and prompts for clarity, consistency, and performance improvements
- feat: optimize ANSI syntax highlighting in diff renderer for improved performance
- feat: add agent option to CLI for temporary model override
- feat: enhance planning tool with quality validation and detailed descriptions for task phases
- feat: enhance input history navigation and improve session input handling
- feat: enhance tool execution logging and improve diff preview generation
- feat: improve error handling for create_file and update_plan methods, enhance logging for theme loading failures
- feat: implement adaptive TUI tick rate, coalesce scroll events, and enhance session management
- feat: implement adaptive TUI tick rate and coalesce scroll events
- feat: Add alias for /config command as /settings, enhance slash command descriptions, and introduce quiet output in configuration
- feat: Improve terminal detection and configuration path resolution across operating systems, update LLM provider integrations, and refine agent slash commands and welcome flow.
- feat: Integrate `TimeoutsConfig` into LLM provider HTTP clients and refactor OpenRouter error handling.
- feat(llm): Introduce a centralized HTTP client factory, refactor providers to use it for consistent timeout configuration, and enhance API error parsing.
- feat: Improve tool input deserialization to handle quoted values, enhance `grep` path validation, and update tool policies.
- feat: Introduce GLM-4.7 Novita model, prepend system prompts in HuggingFace provider, skip GLM thinking parameter, and update tool policies.
- feat: Introduce `--quiet` flag and separate `stdout` for data and `stderr` for logs to improve CLI piping.
- feat: Refactor and expand slash command handling with new diagnostics, skills, tools, workspace, and context commands.
- feat: Implement terminal setup wizard with support for multiple terminals and features, and update LLM provider models.
- feat: add code intelligence tool with LSP-like navigation features
### Bug Fixes
- fix: Disable JSON object output and Responses API for GLM models and refine streaming completion event content handling.
- fix: disable npm publishing in release.sh
- fix: remove npm installation due to GitHub Actions costs
### Refactors
- refactor: update reasoning labels for clarity in justification and session headers
- refactor: streamline toolset by merging agent diagnostics and removing deprecated tools
- refactor: simplify error handling in dotenv loading
### Documentation
- docs: update changelog for v0.53.0 [skip ci]
- docs: Update README with new sections for Keyboard Shortcuts and macOS Alt Shortcut Troubleshooting; refine TODO list entries for clarity and consistency.
- docs: Add a comprehensive list of new features, bug fixes, and performance improvements to the project TODO list.
- docs: update changelog for v0.52.10 [skip ci]
- docs: update changelog for v0.52.9 [skip ci]
- docs: update installation instructions and scripts for npm package
### Chores
- chore: remove completed tasks from TODO.md and improve memory usage for large conversations
- chore: update npm package.json to v0.53.0 [skip ci]
- chore: release v0.53.0
- chore: release v0.52.10
- chore: release v0.52.9
- chore(deps): bump the all-rust-deps group with 21 updates
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 21 to 22
- chore(deps): bump actions/cache from 4 to 5
- chore(deps): bump actions/upload-artifact from 5 to 6
# [Version 0.53.0] - 2025-12-25
### Features
- feat: add agent option to CLI for temporary model override
- feat: enhance planning tool with quality validation and detailed descriptions for task phases
- feat: enhance input history navigation and improve session input handling
- feat: enhance tool execution logging and improve diff preview generation
- feat: improve error handling for create_file and update_plan methods, enhance logging for theme loading failures
- feat: implement adaptive TUI tick rate, coalesce scroll events, and enhance session management
- feat: implement adaptive TUI tick rate and coalesce scroll events
- feat: Add alias for /config command as /settings, enhance slash command descriptions, and introduce quiet output in configuration
- feat: Improve terminal detection and configuration path resolution across operating systems, update LLM provider integrations, and refine agent slash commands and welcome flow.
- feat: Integrate `TimeoutsConfig` into LLM provider HTTP clients and refactor OpenRouter error handling.
- feat(llm): Introduce a centralized HTTP client factory, refactor providers to use it for consistent timeout configuration, and enhance API error parsing.
- feat: Improve tool input deserialization to handle quoted values, enhance `grep` path validation, and update tool policies.
- feat: Introduce GLM-4.7 Novita model, prepend system prompts in HuggingFace provider, skip GLM thinking parameter, and update tool policies.
- feat: Introduce `--quiet` flag and separate `stdout` for data and `stderr` for logs to improve CLI piping.
- feat: Refactor and expand slash command handling with new diagnostics, skills, tools, workspace, and context commands.
- feat: Implement terminal setup wizard with support for multiple terminals and features, and update LLM provider models.
- feat: add code intelligence tool with LSP-like navigation features
### Bug Fixes
- fix: Disable JSON object output and Responses API for GLM models and refine streaming completion event content handling.
- fix: disable npm publishing in release.sh
- fix: remove npm installation due to GitHub Actions costs
- fix: rename npm package from vtcode-bin to vtcode
### Documentation
- docs: Update README with new sections for Keyboard Shortcuts and macOS Alt Shortcut Troubleshooting; refine TODO list entries for clarity and consistency.
- docs: Add a comprehensive list of new features, bug fixes, and performance improvements to the project TODO list.
- docs: update changelog for v0.52.10 [skip ci]
- docs: update changelog for v0.52.9 [skip ci]
- docs: update installation instructions and scripts for npm package
### Chores
- chore: release v0.52.10
- chore: release v0.52.9
- chore(deps): bump the all-rust-deps group with 21 updates
- chore: release v0.52.8
- chore: update npm version to 0.52.8
- chore: release v0.52.7
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 21 to 22
- chore(deps): bump actions/cache from 4 to 5
- chore(deps): bump actions/upload-artifact from 5 to 6
# [Version 0.52.10] - 2025-12-25
### Features
- feat: Refactor and expand slash command handling with new diagnostics, skills, tools, workspace, and context commands.
- feat: Implement terminal setup wizard with support for multiple terminals and features, and update LLM provider models.
- feat: add code intelligence tool with LSP-like navigation features
### Bug Fixes
- fix: Disable JSON object output and Responses API for GLM models and refine streaming completion event content handling.
- fix: disable npm publishing in release.sh
- fix: remove npm installation due to GitHub Actions costs
- fix: rename npm package from vtcode-bin to vtcode
### Documentation
- docs: update changelog for v0.52.9 [skip ci]
- docs: update installation instructions and scripts for npm package
### Chores
- chore: release v0.52.9
- chore(deps): bump the all-rust-deps group with 21 updates
- chore: release v0.52.8
- chore: update npm version to 0.52.8
- chore: release v0.52.7
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 21 to 22
- chore(deps): bump actions/cache from 4 to 5
- chore(deps): bump actions/upload-artifact from 5 to 6
# [Version 0.52.9] - 2025-12-25
### Features
- feat: Refactor and expand slash command handling with new diagnostics, skills, tools, workspace, and context commands.
- feat: Implement terminal setup wizard with support for multiple terminals and features, and update LLM provider models.
- feat: add code intelligence tool with LSP-like navigation features
### Bug Fixes
- fix: disable npm publishing in release.sh
- fix: remove npm installation due to GitHub Actions costs
- fix: rename npm package from vtcode-bin to vtcode
### Documentation
- docs: update installation instructions and scripts for npm package
### Chores
- chore(deps): bump the all-rust-deps group with 21 updates
- chore: release v0.52.8
- chore: update npm version to 0.52.8
- chore: release v0.52.7
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 21 to 22
- chore(deps): bump actions/cache from 4 to 5
- chore(deps): bump actions/upload-artifact from 5 to 6
# [Version 0.52.5] - 2025-12-24
### Bug Fixes
- fix: update release workflow to handle npm publishing correctly
- fix: unignore .github directory to enable GitHub Actions CI/CD workflows
### Documentation
- docs: update changelog for v0.52.4 [skip ci]
### Chores
- chore: update npm package.json to v0.52.4 [skip ci]
- chore: release v0.52.4
- chore: update npm package.json to v0.52.3 [skip ci]
# [Version 0.52.4] - 2025-12-24
### Features
- feat: Add new Z.AI GLM models, refine reasoning support, and update Hugging Face model naming conventions.
- feat: reimplement HuggingFace LLM provider with dedicated logic to handle its unique API behaviors and compatibility.
- feat: Add Hugging Face integration documentation and update tool policies to include git and cargo commands while removing some mcp time-related tools.
### Bug Fixes
- fix: unignore .github directory to enable GitHub Actions CI/CD workflows
### Documentation
- docs: update changelog for v0.52.3 [skip ci]
### Chores
- chore: update npm package.json to v0.52.3 [skip ci]
- chore: release v0.52.3
- chore: update npm package.json to v0.52.2 [skip ci]
# [Version 0.52.3] - 2025-12-24
### Features
- feat: Add new Z.AI GLM models, refine reasoning support, and update Hugging Face model naming conventions.
- feat: reimplement HuggingFace LLM provider with dedicated logic to handle its unique API behaviors and compatibility.
- feat: Add Hugging Face integration documentation and update tool policies to include git and cargo commands while removing some mcp time-related tools.
- feat: Add MiniMax model support to the Anthropic provider and adjust its API base URL.
- feat: Reorganize Hugging Face model identifiers and enhance Anthropic model validation
- feat: Add Hugging Face provider support and update configuration
- feat: Update model provider to OpenAI and enhance Responses API handling
- feat: Expand Hugging Face model support and update provider implementation
- feat: Update model references and configuration for Z.AI GLM-4.7
- feat: Add missing OpenRouter model entries and update reasoning handling
- feat: Enhance OpenAI responses handling with tool call parsing and sampling parameters
- feat: Update tool policies and add new Grok models to configuration
- feat: Add Z.AI GLM-4.7 model to models.json and update constants
- feat: Include Claude agent configurations and GitHub workflows in version control, and update existing agent definitions, skills, commands, hooks, and CI/CD configurations.
- feat: Introduce agent giving-up reasoning detection and constructive responses, and set `execute_code` tool policy to prompt.
### Bug Fixes
- fix: add missing package-lock.json r npm CI workflow
### Documentation
- docs: update changelog for v0.52.2 [skip ci]
### Chores
- chore: update npm package.json to v0.52.2 [skip ci]
- chore: release v0.52.2
- chore: remove temporary file `temp_check.rs`
- chore: update npm package.json to v0.52.1 [skip ci]
# [Version 0.52.2] - 2025-12-24
### Features
- feat: Add MiniMax model support to the Anthropic provider and adjust its API base URL.
- feat: Reorganize Hugging Face model identifiers and enhance Anthropic model validation
- feat: Add Hugging Face provider support and update configuration
- feat: Update model provider to OpenAI and enhance Responses API handling
- feat: Expand Hugging Face model support and update provider implementation
- feat: Update model references and configuration for Z.AI GLM-4.7
- feat: Add missing OpenRouter model entries and update reasoning handling
- feat: Enhance OpenAI responses handling with tool call parsing and sampling parameters
- feat: Update tool policies and add new Grok models to configuration
- feat: Add Z.AI GLM-4.7 model to models.json and update constants
- feat: Include Claude agent configurations and GitHub workflows in version control, and update existing agent definitions, skills, commands, hooks, and CI/CD configurations.
- feat: Introduce agent giving-up reasoning detection and constructive responses, and set `execute_code` tool policy to prompt.
- feat: add keyboard protocol configuration and documentation for enhanced keyboard event handling
### Bug Fixes
- fix: add missing package-lock.json r npm CI workflow
### Refactors
- refactor: simplify configuration handling and update tool permissions in multiple files
- refactor: apply clippy fixes for code quality improvements
- refactor: clean up whitespace and formatting across multiple files for improved readability
### Documentation
- docs: update changelog for v0.52.1 [skip ci]
### Tests
- test: add missing fields to LLMRequest initializers
### Chores
- chore: remove temporary file `temp_check.rs`
- chore: update npm package.json to v0.52.1 [skip ci]
- chore: release v0.52.1
- chore: update npm package.json to v0.52.0 [skip ci]
# [Version 0.52.1] - 2025-12-23
### Features
- feat: add keyboard protocol configuration and documentation for enhanced keyboard event handling
- feat: Introduce advanced LLM parameters, add default implementations for LLMRequest, Message, and ToolChoice, and remove nextest.toml.
### Refactors
- refactor: simplify configuration handling and update tool permissions in multiple files
- refactor: apply clippy fixes for code quality improvements
- refactor: clean up whitespace and formatting across multiple files for improved readability
- refactor: Migrate testing from `cargo nextest` to `cargo test` and enhance Anthropic LLM configuration with new parameters.
### Documentation
- docs: update changelog for v0.52.0 [skip ci]
- docs: Add guidelines for git operations in AGENTS.md and update LLM provider configuration in vtcode.toml
- docs: Replace all cargo nextest references with cargo test across documentation and agent rules
### Tests
- test: add missing fields to LLMRequest initializers
### Chores
- chore: update npm package.json to v0.52.0 [skip ci]
- chore: release v0.52.0
- chore: update npm package.json to v0.51.2 [skip ci]
# [Version 0.52.0] - 2025-12-23
### Features
- feat: Introduce advanced LLM parameters, add default implementations for LLMRequest, Message, and ToolChoice, and remove nextest.toml.
- feat: Add new model constants for grok-4-1-fast and grok-code-fast-1
### Refactors
- refactor: Migrate testing from `cargo nextest` to `cargo test` and enhance Anthropic LLM configuration with new parameters.
### Documentation
- docs: Add guidelines for git operations in AGENTS.md and update LLM provider configuration in vtcode.toml
- docs: Replace all cargo nextest references with cargo test across documentation and agent rules
- docs: update changelog for v0.51.2 [skip ci]
### Chores
- chore: update npm package.json to v0.51.2 [skip ci]
- chore: release v0.51.2
- chore: update npm package.json to v0.51.1 [skip ci]
# [Version 0.51.2] - 2025-12-22
### Features
- feat: Add new model constants for grok-4-1-fast and grok-code-fast-1
- feat: Implement search and filter functionality for the TUI configuration palette.
- feat: Introduce a TUI config palette, refactor rendering logic, and enable dynamic theme application.
### Documentation
- docs: update changelog for v0.51.1 [skip ci]
- docs: update changelog for v0.51.0 [skip ci]
### Chores
- chore: update npm package.json to v0.51.1 [skip ci]
- chore: release v0.51.1
- chore: release v0.51.0
- chore: update npm package.json to v0.50.13 [skip ci]
# [Version 0.51.1] - 2025-12-22
### Features
- feat: Implement search and filter functionality for the TUI configuration palette.
- feat: Introduce a TUI config palette, refactor rendering logic, and enable dynamic theme application.
- feat: enhance session logging functionality and update default model
- feat: integrate SessionWidget into main render function
- feat: add buffer-based widgets for input, modal, and slash
- feat: create ratatui widget foundation
- feat: implement centralized panic handling for TUI applications
- feat: add better panic handling with debug mode support
- feat: enhance list rendering with highlight symbol and repeat option
- feat: implement XDG Base Directory Specification for configuration and data storage
### Bug Fixes
- fix: redirect terminal commands from stdout to stderr for TUI functionality
- fix: change terminal output from stderr to stdout for ModernTui
- fix: reorder MCP time policies and update tool policy documentation
- fix: align OpenAI Responses API implementation with official spec
### Refactors
- refactor: streamline widget block creation and layout definitions in TUI components
### Documentation
- docs: update changelog for v0.51.0 [skip ci]
- docs: update changelog for v0.50.13 [skip ci]
### Chores
- chore: release v0.51.0
- chore: update npm package.json to v0.50.13 [skip ci]
- chore: release v0.50.13
- chore: update npm package.json to v0.50.12 [skip ci]
# [Version 0.51.0] - 2025-12-22
### Features
- feat: Introduce a TUI config palette, refactor rendering logic, and enable dynamic theme application.
- feat: enhance session logging functionality and update default model
- feat: integrate SessionWidget into main render function
- feat: add buffer-based widgets for input, modal, and slash
- feat: create ratatui widget foundation
- feat: implement centralized panic handling for TUI applications
- feat: add better panic handling with debug mode support
- feat: enhance list rendering with highlight symbol and repeat option
- feat: implement XDG Base Directory Specification for configuration and data storage
### Bug Fixes
- fix: redirect terminal commands from stdout to stderr for TUI functionality
- fix: change terminal output from stderr to stdout for ModernTui
- fix: reorder MCP time policies and update tool policy documentation
- fix: align OpenAI Responses API implementation with official spec
### Refactors
- refactor: streamline widget block creation and layout definitions in TUI components
### Documentation
- docs: update changelog for v0.50.13 [skip ci]
### Chores
- chore: update npm package.json to v0.50.13 [skip ci]
- chore: release v0.50.13
- chore: update npm package.json to v0.50.12 [skip ci]
# [Version 0.50.13] - 2025-12-21
### Features
- feat: enhance session logging functionality and update default model
- feat: integrate SessionWidget into main render function
- feat: add buffer-based widgets for input, modal, and slash
- feat: create ratatui widget foundation
- feat: implement centralized panic handling for TUI applications
- feat: add better panic handling with debug mode support
- feat: enhance list rendering with highlight symbol and repeat option
- feat: implement XDG Base Directory Specification for configuration and data storage
- feat: Enhance tool execution error handling and implement Planning workflow warnings
- feat: Implement TUI-aware tool approval prompts and human-in-the-loop notification bell.
- feat: add plan phase management and update tool registry for the Planning workflow
- feat: add HITL notification bell configuration and implement terminal bell notification for approvals
- feat: Implement pre-flight LLM request and tool definition validation, and ensure `mark_tool_loop_limit_hit` is idempotent.
- feat: improve release process by adding Linux build automation and related documentation.
### Bug Fixes
- fix: redirect terminal commands from stdout to stderr for TUI functionality
- fix: change terminal output from stderr to stdout for ModernTui
- fix: reorder MCP time policies and update tool policy documentation
- fix: align OpenAI Responses API implementation with official spec
### Performance Improvements
- perf: optimize rate limiting with a read-lock fast path and refactor tool execution retry delays using constant values.
### Refactors
- refactor: streamline widget block creation and layout definitions in TUI components
### Documentation
- docs: update changelog for v0.50.12 [skip ci]
### Chores
- chore: update npm package.json to v0.50.12 [skip ci]
- chore: release v0.50.12
- chore: update npm package.json to v0.50.11 [skip ci]
# [Version 0.50.12] - 2025-12-20
### Features
- feat: Enhance tool execution error handling and implement Planning workflow warnings
- feat: Implement TUI-aware tool approval prompts and human-in-the-loop notification bell.
- feat: add plan phase management and update tool registry for the Planning workflow
- feat: add HITL notification bell configuration and implement terminal bell notification for approvals
- feat: Implement pre-flight LLM request and tool definition validation, and ensure `mark_tool_loop_limit_hit` is idempotent.
- feat: improve release process by adding Linux build automation and related documentation.
- feat: Add Linux build and release support, fix npm publish, and improve release asset uploads and install script error handling.
### Performance Improvements
- perf: optimize rate limiting with a read-lock fast path and refactor tool execution retry delays using constant values.
### Documentation
- docs: update changelog for v0.50.11 [skip ci]
### Chores
- chore: update npm package.json to v0.50.11 [skip ci]
- chore: release v0.50.11
- chore: update VSCode extension package.json to v0.50.10 [skip ci]
- chore: update npm package.json to v0.50.10 [skip ci]
# [Version 0.50.11] - 2025-12-20
### Features
- feat: Add Linux build and release support, fix npm publish, and improve release asset uploads and install script error handling.
- feat: Add npm publishing troubleshooting guide and authentication setup script, and automate binary stub creation in the release process.
- feat: Enable manual versioned builds in the release workflow and significantly enhance the install script with improved dependency/platform detection, asset verification, and a cargo fallback.
### Bug Fixes
- fix: resolve GitHub release binary upload failures and enhance release script verification with new documentation.
### Documentation
- docs: update changelog for v0.50.10 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.10 [skip ci]
- chore: update npm package.json to v0.50.10 [skip ci]
- chore: release v0.50.10
- chore: update VSCode extension package.json to v0.50.9 [skip ci]
- chore: update npm package.json to v0.50.9 [skip ci]
# [Version 0.50.10] - 2025-12-20
### Features
- feat: Add npm publishing troubleshooting guide and authentication setup script, and automate binary stub creation in the release process.
- feat: Enable manual versioned builds in the release workflow and significantly enhance the install script with improved dependency/platform detection, asset verification, and a cargo fallback.
- feat: Implement agent task retry with exponential backoff and render tool follow-up prompts.
- feat: Enhance skill validation and file reference checks for Agent Skills compliance
- feat: Implement Agent Skills specification by adding `compatibility` and `metadata` fields to skill manifests, updating `allowed-tools` to a space-delimited string, and clarifying skill loading behavior.
- feat: Enhance tool execution policy with granular user confirmation, auto-acceptance, and feedback capabilities.
- feat: extract anthropic config, reduce configuration complexity, document experimental features
- feat: remove reinforcement learning and optimization modules and configurations.
- feat: Add context-aware prompt enrichment (vibe coding) with new context modules and wizard modal interaction events.
- feat: Add full-auto permission review support and update tool policies for improved automation
- feat: Implement per-tool rate limiting and refactor agent tool execution state management.
- feat: implement circuit breaker pattern for MCP client failures and optimize tool inventory management
- feat: add code reviewer and commit message generator skills
### Bug Fixes
- fix: resolve GitHub release binary upload failures and enhance release script verification with new documentation.
- fix: update tool policies and configuration settings for improved performance
- fix: remove external editor keybinding (Control+E)
- fix: prevent arrow keys from triggering external editor launch
### Refactors
- refactor: remove router configuration and related core logic
### Documentation
- docs: update changelog for v0.50.9 [skip ci]
- docs: update changelog for v0.50.8 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.9 [skip ci]
- chore: update npm package.json to v0.50.9 [skip ci]
- chore: release v0.50.9
- chore: release v0.50.8
- chore: remove AI model routing configuration from TOML files
- chore: update VSCode extension package.json to v0.50.7 [skip ci]
- chore: update npm package.json to v0.50.7 [skip ci]
# [Version 0.50.9] - 2025-12-20
### Features
- feat: Implement agent task retry with exponential backoff and render tool follow-up prompts.
- feat: Enhance skill validation and file reference checks for Agent Skills compliance
- feat: Implement Agent Skills specification by adding `compatibility` and `metadata` fields to skill manifests, updating `allowed-tools` to a space-delimited string, and clarifying skill loading behavior.
- feat: Enhance tool execution policy with granular user confirmation, auto-acceptance, and feedback capabilities.
- feat: extract anthropic config, reduce configuration complexity, document experimental features
- feat: remove reinforcement learning and optimization modules and configurations.
- feat: Add context-aware prompt enrichment (vibe coding) with new context modules and wizard modal interaction events.
- feat: Add full-auto permission review support and update tool policies for improved automation
- feat: Implement per-tool rate limiting and refactor agent tool execution state management.
- feat: implement circuit breaker pattern for MCP client failures and optimize tool inventory management
- feat: add code reviewer and commit message generator skills
### Bug Fixes
- fix: update tool policies and configuration settings for improved performance
- fix: remove external editor keybinding (Control+E)
- fix: prevent arrow keys from triggering external editor launch
### Refactors
- refactor: remove router configuration and related core logic
### Documentation
- docs: update changelog for v0.50.8 [skip ci]
- docs: update changelog for v0.50.7 [skip ci]
- docs: update changelog for v0.50.6 [skip ci]
### Chores
- chore: release v0.50.8
- chore: remove AI model routing configuration from TOML files
- chore: update VSCode extension package.json to v0.50.7 [skip ci]
- chore: update npm package.json to v0.50.7 [skip ci]
- chore: release v0.50.7
- chore: release v0.50.6
- chore: update VSCode extension package.json to v0.50.5 [skip ci]
- chore: update npm package.json to v0.50.5 [skip ci]
# [Version 0.50.8] - 2025-12-20
### Features
- feat: Implement Agent Skills specification by adding `compatibility` and `metadata` fields to skill manifests, updating `allowed-tools` to a space-delimited string, and clarifying skill loading behavior.
- feat: Enhance tool execution policy with granular user confirmation, auto-acceptance, and feedback capabilities.
- feat: extract anthropic config, reduce configuration complexity, document experimental features
- feat: remove reinforcement learning and optimization modules and configurations.
- feat: Add context-aware prompt enrichment (vibe coding) with new context modules and wizard modal interaction events.
- feat: Add full-auto permission review support and update tool policies for improved automation
- feat: Implement per-tool rate limiting and refactor agent tool execution state management.
- feat: implement circuit breaker pattern for MCP client failures and optimize tool inventory management
- feat: add code reviewer and commit message generator skills
### Bug Fixes
- fix: update tool policies and configuration settings for improved performance
- fix: remove external editor keybinding (Control+E)
- fix: prevent arrow keys from triggering external editor launch
### Refactors
- refactor: remove router configuration and related core logic
### Documentation
- docs: update changelog for v0.50.7 [skip ci]
- docs: update changelog for v0.50.6 [skip ci]
### Chores
- chore: remove AI model routing configuration from TOML files
- chore: update VSCode extension package.json to v0.50.7 [skip ci]
- chore: update npm package.json to v0.50.7 [skip ci]
- chore: release v0.50.7
- chore: release v0.50.6
- chore: update VSCode extension package.json to v0.50.5 [skip ci]
- chore: update npm package.json to v0.50.5 [skip ci]
# [Version 0.50.7] - 2025-12-19
### Features
- feat: Add success indicators to renderer on exit commands and session end
- feat: Update tool policy to prompt for file creation and execution, enhance session state management, and adjust LLM provider settings in configuration.
- feat: Enhance skill definitions with new metadata fields, improve LLM provider support, and refine TUI components.
- feat: Implement shell command policy checking with regex and glob patterns and add new metadata fields to skill definitions.
- feat: enhance line ending handling in patch operations and tests
### Bug Fixes
- fix: Correct test expectations for token threshold boundaries
- fix: handle errors in AtomicWriter creation and improve diff operations tests
### Refactors
- refactor: remove unnecessary whitespace in diff and test files
### Documentation
- docs: update changelog for v0.50.6 [skip ci]
- docs: update changelog for v0.50.5 [skip ci]
### Chores
- chore: release v0.50.6
- chore: update VSCode extension package.json to v0.50.5 [skip ci]
- chore: update npm package.json to v0.50.5 [skip ci]
- chore: release v0.50.5
- chore: update VSCode extension package.json to v0.50.4 [skip ci]
- chore: update npm package.json to v0.50.4 [skip ci]
# [Version 0.50.6] - 2025-12-19
### Features
- feat: Add success indicators to renderer on exit commands and session end
- feat: Update tool policy to prompt for file creation and execution, enhance session state management, and adjust LLM provider settings in configuration.
- feat: Enhance skill definitions with new metadata fields, improve LLM provider support, and refine TUI components.
- feat: Implement shell command policy checking with regex and glob patterns and add new metadata fields to skill definitions.
- feat: enhance line ending handling in patch operations and tests
### Bug Fixes
- fix: Correct test expectations for token threshold boundaries
- fix: handle errors in AtomicWriter creation and improve diff operations tests
### Refactors
- refactor: remove unnecessary whitespace in diff and test files
### Documentation
- docs: update changelog for v0.50.5 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.5 [skip ci]
- chore: update npm package.json to v0.50.5 [skip ci]
- chore: release v0.50.5
- chore: update VSCode extension package.json to v0.50.4 [skip ci]
- chore: update npm package.json to v0.50.4 [skip ci]
# [Version 0.50.5] - 2025-12-19
### Features
- feat: Add success indicators to renderer on exit commands and session end
- feat: Update tool policy to prompt for file creation and execution, enhance session state management, and adjust LLM provider settings in configuration.
- feat: Enhance skill definitions with new metadata fields, improve LLM provider support, and refine TUI components.
- feat: Implement shell command policy checking with regex and glob patterns and add new metadata fields to skill definitions.
- feat: enhance line ending handling in patch operations and tests
- feat: add Gemini 3 Flash Preview model and update configurations
### Bug Fixes
- fix: Correct test expectations for token threshold boundaries
- fix: handle errors in AtomicWriter creation and improve diff operations tests
### Refactors
- refactor: remove unnecessary whitespace in diff and test files
### Documentation
- docs: update changelog for v0.50.4 [skip ci]
- docs: update changelog for v0.50.3 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.4 [skip ci]
- chore: update npm package.json to v0.50.4 [skip ci]
- chore: release v0.50.4
- chore: release v0.50.3
- chore: update VSCode extension package.json to v0.50.2 [skip ci]
- chore: update npm package.json to v0.50.2 [skip ci]
# [Version 0.50.4] - 2025-12-18
### Features
- feat: add Gemini 3 Flash Preview model and update configurations
### Documentation
- docs: update changelog for v0.50.3 [skip ci]
- docs: update changelog for v0.50.2 [skip ci]
### Chores
- chore: release v0.50.3
- chore: update VSCode extension package.json to v0.50.2 [skip ci]
- chore: update npm package.json to v0.50.2 [skip ci]
- chore: release v0.50.2
- chore: update VSCode extension package.json to v0.50.1 [skip ci]
- chore: update npm package.json to v0.50.1 [skip ci]
# [Version 0.50.3] - 2025-12-18
### Features
- feat: add Gemini 3 Flash Preview model and update configurations
### Documentation
- docs: update changelog for v0.50.2 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.2 [skip ci]
- chore: update npm package.json to v0.50.2 [skip ci]
- chore: release v0.50.2
- chore: update VSCode extension package.json to v0.50.1 [skip ci]
- chore: update npm package.json to v0.50.1 [skip ci]
# [Version 0.50.2] - 2025-12-16
### Bug Fixes
- fix: correct logical operator for XAI provider model check
### Documentation
- docs: update changelog for v0.50.1 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.1 [skip ci]
- chore: update npm package.json to v0.50.1 [skip ci]
- chore: release v0.50.1
- chore: update VSCode extension package.json to v0.50.0 [skip ci]
- chore: update npm package.json to v0.50.0 [skip ci]
# [Version 0.50.1] - 2025-12-16
### Features
- feat: implement rate limiting for tool calls and add Nemotron-3-Nano model support
### Bug Fixes
- fix: correct logical operator for XAI provider model check
### Documentation
- docs: update changelog for v0.50.0 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.50.0 [skip ci]
- chore: update npm package.json to v0.50.0 [skip ci]
- chore: release v0.50.0
- chore: update VSCode extension package.json to v0.49.8 [skip ci]
- chore: update npm package.json to v0.49.8 [skip ci]
# [Version 0.50.0] - 2025-12-16
### Features
- feat: implement rate limiting for tool calls and add Nemotron-3-Nano model support
### Documentation
- docs: update changelog for v0.49.8 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.49.8 [skip ci]
- chore: update npm package.json to v0.49.8 [skip ci]
- chore: release v0.49.8
- chore: update VSCode extension package.json to v0.49.7 [skip ci]
- chore: update npm package.json to v0.49.7 [skip ci]
# [Version 0.49.8] - 2025-12-16
### Documentation
- docs: update changelog for v0.49.7 [skip ci]
- docs: update changelog for v0.49.6 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.49.7 [skip ci]
- chore: update npm package.json to v0.49.7 [skip ci]
- chore: release v0.49.7
- chore: release v0.49.6
- chore: update VSCode extension package.json to v0.49.5 [skip ci]
- chore: update npm package.json to v0.49.5 [skip ci]
# [Version 0.49.7] - 2025-12-15
### Bug Fixes
- fix: include templates directory in package for crates.io publishing
### Documentation
- docs: update changelog for v0.49.6 [skip ci]
- docs: update changelog for v0.49.5 [skip ci]
- docs: update changelog for v0.49.4 [skip ci]
- docs: update changelog for v0.49.3 [skip ci]
- docs: update changelog for v0.49.2 [skip ci]
### Chores
- chore: release v0.49.6
- chore: update VSCode extension package.json to v0.49.5 [skip ci]
- chore: update npm package.json to v0.49.5 [skip ci]
- chore: release v0.49.5
- chore: release v0.49.4
- chore: release v0.49.3
- chore: release v0.49.2
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
# [Version 0.49.6] - 2025-12-15
### Bug Fixes
- fix: include templates directory in package for crates.io publishing
### Documentation
- docs: update changelog for v0.49.5 [skip ci]
- docs: update changelog for v0.49.4 [skip ci]
- docs: update changelog for v0.49.3 [skip ci]
- docs: update changelog for v0.49.2 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.49.5 [skip ci]
- chore: update npm package.json to v0.49.5 [skip ci]
- chore: release v0.49.5
- chore: release v0.49.4
- chore: release v0.49.3
- chore: release v0.49.2
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
# [Version 0.49.5] - 2025-12-14
### Bug Fixes
- fix: include templates directory in package for crates.io publishing
### Documentation
- docs: update changelog for v0.49.4 [skip ci]
- docs: update changelog for v0.49.3 [skip ci]
- docs: update changelog for v0.49.2 [skip ci]
- docs: update changelog for v0.49.1 [skip ci]
### Chores
- chore: release v0.49.4
- chore: release v0.49.3
- chore: release v0.49.2
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
- chore: release v0.49.1
- chore: update npm package.json to v0.49.0 [skip ci]
# [Version 0.49.4] - 2025-12-14
### Documentation
- docs: update changelog for v0.49.3 [skip ci]
- docs: update changelog for v0.49.2 [skip ci]
- docs: update changelog for v0.49.1 [skip ci]
### Chores
- chore: release v0.49.3
- chore: release v0.49.2
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
- chore: release v0.49.1
- chore: update npm package.json to v0.49.0 [skip ci]
# [Version 0.49.3] - 2025-12-14
### Documentation
- docs: update changelog for v0.49.2 [skip ci]
- docs: update changelog for v0.49.1 [skip ci]
### Chores
- chore: release v0.49.2
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
- chore: release v0.49.1
- chore: update npm package.json to v0.49.0 [skip ci]
# [Version 0.49.2] - 2025-12-14
### Documentation
- docs: update changelog for v0.49.1 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.49.1 [skip ci]
- chore: update npm package.json to v0.49.1 [skip ci]
- chore: release v0.49.1
- chore: update npm package.json to v0.49.0 [skip ci]
### Added
- **Comprehensive Skills Location System**: Implemented multi-location skill discovery with precedence handling
- VT Code User Skills (`~/.vtcode/skills/`) - Highest precedence
- VT Code Project Skills (`.vtcode/skills/`) - Project-specific skills
- Pi Framework Skills (`~/.pi/skills/`, `.pi/skills/`)
- Claude Code Skills (`~/.claude/skills/`, `.claude/skills/`)
- Codex CLI Skills (`~/.codex/skills/`)
- **Precedence System**: Skills from higher precedence locations override lower precedence skills with the same name
- **Migration Support**: All existing skills migrated from `.claude/skills` to `.vtcode/skills` with backward compatibility
- **Enhanced Skill Loader**: Updated loader to integrate with new location system while maintaining backward compatibility
### Changed
- Updated skills documentation to reflect new location system and precedence handling
- Enhanced skill discovery to support recursive scanning and proper name collision resolution
## [Version 0.43.0] - 2025-11-09
# [Version 0.49.1] - 2025-12-13
### Bug Fixes
- fix: update execute_code and skill policies to allow execution
- fix: remove outdated skill discovery documentation and integrate new skill loading functionality
- fix: enhance skill discovery and loading functionality for vtcode agent
- fix: improve skill tool output to include full instructions
- fix: add missing skill tool function declaration
- fix: vtcode agent skill discovery using SkillLoader instead of SkillManager
### Documentation
- docs: update changelog for v0.49.0 [skip ci]
- docs: add complete skill tool fix summary
### Chores
- chore: update npm package.json to v0.49.0 [skip ci]
- chore: release v0.49.0
- chore: update VSCode extension package.json to v0.48.3 [skip ci]
- chore: update npm package.json to v0.48.3 [skip ci]
# [Version 0.49.0] - 2025-12-13
### Bug Fixes
- fix: update execute_code and skill policies to allow execution
- fix: remove outdated skill discovery documentation and integrate new skill loading functionality
- fix: enhance skill discovery and loading functionality for vtcode agent
- fix: improve skill tool output to include full instructions
- fix: add missing skill tool function declaration
- fix: vtcode agent skill discovery using SkillLoader instead of SkillManager
### Documentation
- docs: add complete skill tool fix summary
- docs: update changelog for v0.48.3 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.48.3 [skip ci]
- chore: update npm package.json to v0.48.3 [skip ci]
- chore: release v0.48.3
- chore: update VSCode extension package.json to v0.48.2 [skip ci]
- chore: update npm package.json to v0.48.2 [skip ci]
# [Version 0.48.3] - 2025-12-13
### Refactors
- refactor: update LLMError handling in turn_processing
- refactor: standardize LLMError structure across providers
- refactor: enhance Z.AI provider error handling and API key validation
- refactor: enhance context trimming and session management features
- refactor: update configuration and documentation for improved clarity and performance
### Documentation
- docs: update changelog for v0.48.2 [skip ci]
- docs: update changelog for v0.48.1 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.48.2 [skip ci]
- chore: update npm package.json to v0.48.2 [skip ci]
- chore: release v0.48.2
- chore: release v0.48.1
- chore: update VSCode extension package.json to v0.48.0 [skip ci]
- chore: update npm package.json to v0.48.0 [skip ci]
# [Version 0.48.2] - 2025-12-13
### Features
- feat: update OpenAI provider to support GPT-5.2 and enhance reasoning options
- feat: enhance response output with reasoning traces
- feat: implement timeout warning management for tool execution
- feat: enhance tool execution with rate limiting and workspace management
### Bug Fixes
- fix: improve error handling in LLM client creation and tool execution
### Refactors
- refactor: update LLMError handling in turn_processing
- refactor: standardize LLMError structure across providers
- refactor: enhance Z.AI provider error handling and API key validation
- refactor: enhance context trimming and session management features
- refactor: update configuration and documentation for improved clarity and performance
- refactor: update GPT-5.2 model identifiers and documentation
- refactor: streamline error handling and conditional checks in various modules
- refactor: remove logging statements from UI interaction and session handling
- refactor: remove outdated vibe_tooling_mapping documentation
### Documentation
- docs: update changelog for v0.48.1 [skip ci]
- docs: update changelog for v0.48.0 [skip ci]
### Chores
- chore: release v0.48.1
- chore: update VSCode extension package.json to v0.48.0 [skip ci]
- chore: update npm package.json to v0.48.0 [skip ci]
- chore: release v0.48.0
- chore: update VSCode extension package.json to v0.47.16 [skip ci]
- chore: update npm package.json to v0.47.16 [skip ci]
# [Version 0.48.1] - 2025-12-13
### Features
- feat: update OpenAI provider to support GPT-5.2 and enhance reasoning options
- feat: enhance response output with reasoning traces
- feat: implement timeout warning management for tool execution
- feat: enhance tool execution with rate limiting and workspace management
### Bug Fixes
- fix: improve error handling in LLM client creation and tool execution
### Refactors
- refactor: standardize LLMError structure across providers
- refactor: enhance Z.AI provider error handling and API key validation
- refactor: enhance context trimming and session management features
- refactor: update configuration and documentation for improved clarity and performance
- refactor: update GPT-5.2 model identifiers and documentation
- refactor: streamline error handling and conditional checks in various modules
- refactor: remove logging statements from UI interaction and session handling
- refactor: remove outdated vibe_tooling_mapping documentation
### Documentation
- docs: update changelog for v0.48.0 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.48.0 [skip ci]
- chore: update npm package.json to v0.48.0 [skip ci]
- chore: release v0.48.0
- chore: update VSCode extension package.json to v0.47.16 [skip ci]
- chore: update npm package.json to v0.47.16 [skip ci]
# [Version 0.48.0] - 2025-12-12
### Features
- feat: update OpenAI provider to support GPT-5.2 and enhance reasoning options
- feat: enhance response output with reasoning traces
- feat: implement timeout warning management for tool execution
- feat: enhance tool execution with rate limiting and workspace management
- feat: enhance logging and error handling in orchestrator and agent components
- feat: improve tracing initialization and error handling in main
- feat: implement idle turn detection and management in task execution
- feat: improve HTTP client pool handling and enhance caching middleware
- feat: enhance loop detection and rate limiting in tool execution
- feat: implement timeout management for streaming and generation requests
- feat: implement streaming failure management and cooldown mechanism
- feat: enhance agent logging and improve markdown rendering
- feat: introduce reinforcement learning optimization and enhance configuration
- feat: add new dependencies and improve error handling in main
- feat: enhance grep result optimization and tool registration
- feat: enhance timeout configuration and adaptive timeout handling
- feat: add search_replace tool and enhance file operations
### Bug Fixes
- fix: improve error handling in LLM client creation and tool execution
### Refactors
- refactor: update GPT-5.2 model identifiers and documentation
- refactor: streamline error handling and conditional checks in various modules
- refactor: remove logging statements from UI interaction and session handling
- refactor: remove outdated vibe_tooling_mapping documentation
- refactor: enhance loop detection logic and add tests
### Documentation
- docs: update changelog for v0.47.16 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.16 [skip ci]
- chore: update npm package.json to v0.47.16 [skip ci]
- chore: release v0.47.16
- chore: update VSCode extension package.json to v0.47.15 [skip ci]
- chore: update npm package.json to v0.47.15 [skip ci]
# [Version 0.47.16] - 2025-12-11
### Features
- feat: enhance logging and error handling in orchestrator and agent components
- feat: improve tracing initialization and error handling in main
- feat: implement idle turn detection and management in task execution
- feat: improve HTTP client pool handling and enhance caching middleware
- feat: enhance loop detection and rate limiting in tool execution
- feat: implement timeout management for streaming and generation requests
- feat: implement streaming failure management and cooldown mechanism
- feat: enhance agent logging and improve markdown rendering
- feat: introduce reinforcement learning optimization and enhance configuration
- feat: add new dependencies and improve error handling in main
- feat: enhance grep result optimization and tool registration
- feat: enhance timeout configuration and adaptive timeout handling
- feat: add search_replace tool and enhance file operations
### Refactors
- refactor: enhance loop detection logic and add tests
- refactor: simplify line style selection logic in tool output
- refactor: streamline initialization and error handling in various modules
- refactor: simplify ToolCallUpdateFields initialization
- refactor: update tool policy and improve middleware handling
### Documentation
- docs: update changelog for v0.47.15 [skip ci]
- docs: update changelog for v0.47.14 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.15 [skip ci]
- chore: update npm package.json to v0.47.15 [skip ci]
- chore: release v0.47.15
- chore: release v0.47.14
- chore: update VSCode extension package.json to v0.47.13 [skip ci]
- chore: update npm package.json to v0.47.13 [skip ci]
# [Version 0.47.15] - 2025-12-11
### Refactors
- refactor: simplify line style selection logic in tool output
- refactor: streamline initialization and error handling in various modules
- refactor: simplify ToolCallUpdateFields initialization
- refactor: update tool policy and improve middleware handling
- refactor: remove deprecated tools and update tool policies
- refactor: update tool policy and streamline follow-up handling
- refactor: enhance PTY command output summarization and follow-up handling
- refactor: enhance context management and prompt generation
- refactor: enhance system prompts with improved tool safety and execution guidelines
- refactor: implement tool execution retry logic and enhance error handling
- refactor: enhance tool policy and prompt clarity
- refactor: implement tool denial handling in agent runner
- refactor: enhance McpAllowListConfig structure and improve rule definitions
- refactor: improve code clarity and consistency in multiple modules
- refactor: streamline conditional checks and improve code readability
- refactor: unify token budget constants and improve context management
### Documentation
- docs: update changelog for v0.47.14 [skip ci]
- docs: update changelog for v0.47.13 [skip ci]
### Chores
- chore: release v0.47.14
- chore: update VSCode extension package.json to v0.47.13 [skip ci]
- chore: update npm package.json to v0.47.13 [skip ci]
- chore: release v0.47.13
- chore: update VSCode extension package.json to v0.47.12 [skip ci]
- chore: update npm package.json to v0.47.12 [skip ci]
# [Version 0.47.14] - 2025-12-11
### Refactors
- refactor: simplify line style selection logic in tool output
- refactor: streamline initialization and error handling in various modules
- refactor: simplify ToolCallUpdateFields initialization
- refactor: update tool policy and improve middleware handling
- refactor: remove deprecated tools and update tool policies
- refactor: update tool policy and streamline follow-up handling
- refactor: enhance PTY command output summarization and follow-up handling
- refactor: enhance context management and prompt generation
- refactor: enhance system prompts with improved tool safety and execution guidelines
- refactor: implement tool execution retry logic and enhance error handling
- refactor: enhance tool policy and prompt clarity
- refactor: implement tool denial handling in agent runner
- refactor: enhance McpAllowListConfig structure and improve rule definitions
- refactor: improve code clarity and consistency in multiple modules
- refactor: streamline conditional checks and improve code readability
- refactor: unify token budget constants and improve context management
### Documentation
- docs: update changelog for v0.47.13 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.13 [skip ci]
- chore: update npm package.json to v0.47.13 [skip ci]
- chore: release v0.47.13
- chore: update VSCode extension package.json to v0.47.12 [skip ci]
- chore: update npm package.json to v0.47.12 [skip ci]
# [Version 0.47.13] - 2025-12-10
### Features
- feat: update tool policy and enhance loop detection functionality
- feat: implement tool call safety validation and execution tracking
### Refactors
- refactor: remove deprecated tools and update tool policies
- refactor: update tool policy and streamline follow-up handling
- refactor: enhance PTY command output summarization and follow-up handling
- refactor: enhance context management and prompt generation
- refactor: enhance system prompts with improved tool safety and execution guidelines
- refactor: implement tool execution retry logic and enhance error handling
- refactor: enhance tool policy and prompt clarity
- refactor: implement tool denial handling in agent runner
- refactor: enhance McpAllowListConfig structure and improve rule definitions
- refactor: improve code clarity and consistency in multiple modules
- refactor: streamline conditional checks and improve code readability
- refactor: unify token budget constants and improve context management
- refactor: update tool policy and enhance tool validation
- refactor: improve code formatting and structure across multiple files
- refactor: implement API failure tracking with exponential backoff; optimize tool caching and navigation
- refactor: optimize core agent execution and consolidate utility modules
- refactor: introduce warning and error handling methods in AgentRunner; streamline tool failure logging and path normalization utilities
- refactor: update model configurations to use OpenRouter for Moonshot models; remove deprecated entries and enhance model support
- refactor: enhance reasoning model support and update tool policies; remove deprecated Moonshot models
### Documentation
- docs: update changelog for v0.47.12 [skip ci]
- docs: update changelog for v0.47.11 [skip ci]
- docs: update changelog for v0.47.10 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.12 [skip ci]
- chore: update npm package.json to v0.47.12 [skip ci]
- chore: release v0.47.12
- chore: release v0.47.11
- chore: update dependencies and improve code formatting
- chore: release v0.47.10
- chore: update VSCode extension package.json to v0.47.9 [skip ci]
- chore: update npm package.json to v0.47.9 [skip ci]
# [Version 0.47.12] - 2025-12-08
### Features
- feat: update tool policy and enhance loop detection functionality
- feat: implement tool call safety validation and execution tracking
- feat(build): add multi-stage Dockerfile for building and running vtcode
### Refactors
- refactor: update tool policy and enhance tool validation
- refactor: improve code formatting and structure across multiple files
- refactor: implement API failure tracking with exponential backoff; optimize tool caching and navigation
- refactor: optimize core agent execution and consolidate utility modules
- refactor: introduce warning and error handling methods in AgentRunner; streamline tool failure logging and path normalization utilities
- refactor: update model configurations to use OpenRouter for Moonshot models; remove deprecated entries and enhance model support
- refactor: enhance reasoning model support and update tool policies; remove deprecated Moonshot models
- refactor: improve code readability by simplifying conditional statements and updating deprecated usages across multiple modules
### Documentation
- docs: update changelog for v0.47.11 [skip ci]
- docs: update changelog for v0.47.10 [skip ci]
- docs: update changelog for v0.47.9 [skip ci]
- docs: update changelog for v0.47.8 [skip ci]
### Chores
- chore: release v0.47.11
- chore: update dependencies and improve code formatting
- chore: release v0.47.10
- chore: update VSCode extension package.json to v0.47.9 [skip ci]
- chore: update npm package.json to v0.47.9 [skip ci]
- chore: release v0.47.9
- chore: release v0.47.8
- chore: update tool policies and optimize configuration loading
- chore: update VSCode extension package.json to v0.47.7 [skip ci]
- chore: update npm package.json to v0.47.7 [skip ci]
# [Version 0.47.11] - 2025-12-08
### Features
- feat: update tool policy and enhance loop detection functionality
- feat: implement tool call safety validation and execution tracking
- feat(build): add multi-stage Dockerfile for building and running vtcode
### Refactors
- refactor: update tool policy and enhance tool validation
- refactor: improve code formatting and structure across multiple files
- refactor: implement API failure tracking with exponential backoff; optimize tool caching and navigation
- refactor: optimize core agent execution and consolidate utility modules
- refactor: introduce warning and error handling methods in AgentRunner; streamline tool failure logging and path normalization utilities
- refactor: update model configurations to use OpenRouter for Moonshot models; remove deprecated entries and enhance model support
- refactor: enhance reasoning model support and update tool policies; remove deprecated Moonshot models
- refactor: improve code readability by simplifying conditional statements and updating deprecated usages across multiple modules
### Documentation
- docs: update changelog for v0.47.10 [skip ci]
- docs: update changelog for v0.47.9 [skip ci]
- docs: update changelog for v0.47.8 [skip ci]
### Chores
- chore: update dependencies and improve code formatting
- chore: release v0.47.10
- chore: update VSCode extension package.json to v0.47.9 [skip ci]
- chore: update npm package.json to v0.47.9 [skip ci]
- chore: release v0.47.9
- chore: release v0.47.8
- chore: update tool policies and optimize configuration loading
- chore: update VSCode extension package.json to v0.47.7 [skip ci]
- chore: update npm package.json to v0.47.7 [skip ci]
# [Version 0.47.10] - 2025-12-07
### Features
- feat(build): add multi-stage Dockerfile for building and running vtcode
### Refactors
- refactor: optimize core agent execution and consolidate utility modules
- refactor: introduce warning and error handling methods in AgentRunner; streamline tool failure logging and path normalization utilities
- refactor: update model configurations to use OpenRouter for Moonshot models; remove deprecated entries and enhance model support
- refactor: enhance reasoning model support and update tool policies; remove deprecated Moonshot models
- refactor: improve code readability by simplifying conditional statements and updating deprecated usages across multiple modules
### Documentation
- docs: update changelog for v0.47.9 [skip ci]
- docs: update changelog for v0.47.8 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.9 [skip ci]
- chore: update npm package.json to v0.47.9 [skip ci]
- chore: release v0.47.9
- chore: release v0.47.8
- chore: update tool policies and optimize configuration loading
- chore: update VSCode extension package.json to v0.47.7 [skip ci]
- chore: update npm package.json to v0.47.7 [skip ci]
# [Version 0.47.9] - 2025-12-05
### Features
- feat: update tool policies and improve code structure with dead code allowances
- feat(build): add multi-stage Dockerfile for building and running vtcode
### Refactors
- refactor: improve code readability by simplifying conditional statements and updating deprecated usages across multiple modules
- refactor: remove unused tools from tool policies
### Documentation
- docs: update changelog for v0.47.8 [skip ci]
- docs: update changelog for v0.47.7 [skip ci]
### Chores
- chore: release v0.47.8
- chore: update tool policies and optimize configuration loading
- chore: update VSCode extension package.json to v0.47.7 [skip ci]
- chore: update npm package.json to v0.47.7 [skip ci]
- chore: release v0.47.7
- chore: update VSCode extension package.json to v0.47.6 [skip ci]
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 20 to 21
- chore(deps): bump actions/checkout from 5 to 6
- chore(deps): bump the all-rust-deps group with 15 updates
- chore: update npm package.json to v0.47.6 [skip ci]
# [Version 0.47.8] - 2025-12-05
### Features
- feat: update tool policies and improve code structure with dead code allowances
- feat(build): add multi-stage Dockerfile for building and running vtcode
### Refactors
- refactor: remove unused tools from tool policies
### Documentation
- docs: update changelog for v0.47.7 [skip ci]
### Chores
- chore: update tool policies and optimize configuration loading
- chore: update VSCode extension package.json to v0.47.7 [skip ci]
- chore: update npm package.json to v0.47.7 [skip ci]
- chore: release v0.47.7
- chore: update VSCode extension package.json to v0.47.6 [skip ci]
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 20 to 21
- chore(deps): bump actions/checkout from 5 to 6
- chore(deps): bump the all-rust-deps group with 15 updates
- chore: update npm package.json to v0.47.6 [skip ci]
# [Version 0.47.7] - 2025-12-03
### Features
- feat: update tool policies and improve code structure with dead code allowances
- feat: add diff suppression logic and constants for large changes
### Performance Improvements
- perf: optimize memory allocations and pre-allocate buffers in various modules
- perf: use write! macro in metrics module
- perf: use write! macro in llm/token_metrics
- perf: use write! macro in exec modules and utils
- perf: use write! macro in core token management modules
- perf: use .to_string() instead of format! for context_size
- perf: use write!/writeln! macros in tools and prompts modules
- perf(ui): use write!/writeln! macros in diff_renderer
- perf: optimize string formatting and use unwrap_or_default
- perf: remove unnecessary clone() on Copy types
- perf: use .to_string() directly for Display types instead of .as_str().to_string()
- perf: eliminate redundant clones in config loader
- perf: add Vec with_capacity for known-size allocations
- perf(core): use write! macro instead of format! with push_str
- perf(core): remove redundant clones and allocations
### Refactors
- refactor: remove unused tools from tool policies
- refactor: optimize completion learning modules with .into() patterns
- refactor: optimize linting and code completion modules
- refactor: optimize core modules for code quality and allocations
- refactor: update message type handling and improve system prompt instructions
- refactor(llm): extract serialize_messages_openai_format helper
- refactor(llm): add validate_request_common helper
- refactor(llm): add parse_tool_call and map_finish_reason helpers
- refactor(llm): extract common provider helpers to reduce duplication
- refactor: optimize diff rendering and suppression logic
### Documentation
- docs: update changelog for v0.47.6 [skip ci]
- docs: add comprehensive optimization report
### Style Changes
- style: fix clippy warnings (assign_op, unnecessary_cast, collapsible_if, const thread_local)
- style: remove redundant closures
### Chores
- chore: update VSCode extension package.json to v0.47.6 [skip ci]
- chore(deps): bump DavidAnson/markdownlint-cli2-action from 20 to 21
- chore(deps): bump actions/checkout from 5 to 6
- chore(deps): bump the all-rust-deps group with 15 updates
- chore: update npm package.json to v0.47.6 [skip ci]
- chore: release v0.47.6
- chore: update VSCode extension package.json to v0.47.5 [skip ci]
- chore: update npm package.json to v0.47.5 [skip ci]
# [Version 0.47.6] - 2025-11-30
### Features
- feat: add diff suppression logic and constants for large changes
- feat: parse and display friendly error messages from Anthropic API responses
### Bug Fixes
- fix: clear spinner before displaying error message
- fix: gracefully handle provider API errors without panicking
### Performance Improvements
- perf: optimize memory allocations and pre-allocate buffers in various modules
- perf: use write! macro in metrics module
- perf: use write! macro in llm/token_metrics
- perf: use write! macro in exec modules and utils
- perf: use write! macro in core token management modules
- perf: use .to_string() instead of format! for context_size
- perf: use write!/writeln! macros in tools and prompts modules
- perf(ui): use write!/writeln! macros in diff_renderer
- perf: optimize string formatting and use unwrap_or_default
- perf: remove unnecessary clone() on Copy types
- perf: use .to_string() directly for Display types instead of .as_str().to_string()
- perf: eliminate redundant clones in config loader
- perf: add Vec with_capacity for known-size allocations
- perf(core): use write! macro instead of format! with push_str
- perf(core): remove redundant clones and allocations
### Refactors
- refactor: optimize completion learning modules with .into() patterns
- refactor: optimize linting and code completion modules
- refactor: optimize core modules for code quality and allocations
- refactor: update message type handling and improve system prompt instructions
- refactor(llm): extract serialize_messages_openai_format helper
- refactor(llm): add validate_request_common helper
- refactor(llm): add parse_tool_call and map_finish_reason helpers
- refactor(llm): extract common provider helpers to reduce duplication
- refactor: optimize diff rendering and suppression logic
- refactor: streamline code formatting and improve readability across multiple files
### Documentation
- docs: add comprehensive optimization report
- docs: clarify spinner cleanup implementation details
- docs: update changelog for v0.47.5 [skip ci]
- docs: update changes summary with comprehensive error handling improvements
### Style Changes
- style: fix clippy warnings (assign_op, unnecessary_cast, collapsible_if, const thread_local)
- style: remove redundant closures
### Chores
- chore: update VSCode extension package.json to v0.47.5 [skip ci]
- chore: update npm package.json to v0.47.5 [skip ci]
- chore: release v0.47.5
- chore: update VSCode extension package.json to v0.47.4 [skip ci]
- chore: update npm package.json to v0.47.4 [skip ci]
# [Version 0.47.5] - 2025-11-25
### Features
- feat: parse and display friendly error messages from Anthropic API responses
### Bug Fixes
- fix: clear spinner before displaying error message
- fix: gracefully handle provider API errors without panicking
- fix: remove unused spawn_session import
### Refactors
- refactor: streamline code formatting and improve readability across multiple files
### Documentation
- docs: update changes summary with comprehensive error handling improvements
- docs: update changelog for v0.47.4 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.4 [skip ci]
- chore: update npm package.json to v0.47.4 [skip ci]
- chore: release v0.47.4
- chore: update VSCode extension package.json to v0.47.3 [skip ci]
- chore: update npm package.json to v0.47.3 [skip ci]
# [Version 0.47.4] - 2025-11-25
### Bug Fixes
- fix: remove unused spawn_session import
### Documentation
- docs: update changelog for v0.47.3 [skip ci]
- docs: update changelog for v0.47.2 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.3 [skip ci]
- chore: update npm package.json to v0.47.3 [skip ci]
- chore: release v0.47.3
- chore: release v0.47.2
- chore: update VSCode extension package.json to v0.47.1 [skip ci]
- chore: update npm package.json to v0.47.1 [skip ci]
# [Version 0.47.3] - 2025-11-25
### Features
- feat: Enhance tool policy with pre-approval allowlist, improve file operation error messages, and refine tool declarations.
- feat: Integrate production-grade tool improvements system
- feat: Implement animated thinking spinner for user input submission
- feat: Add comprehensive ANSI escape sequence documentation and a new core utility module for ANSI codes.
### Bug Fixes
- fix: correct RUSTFLAGS invalid option and align system prompt with actual tool definitions
- fix: resolve all cargo clippy warnings and update rust toolchain to stable
- fix: Revise thinking spinner message to use first-person agent voice
- fix: Move thinking spinner display to after user message in transcript
- fix: Clear thinking spinner message on all agent response command types
### Refactors
- refactor: Update ThinkingSpinner struct visibility for better encapsulation
- refactor: remove redundant reasoning handling, clarify intent
### Documentation
- docs: update changelog for v0.47.2 [skip ci]
- docs: update changelog for v0.47.1 [skip ci]
- docs: update changelog for v0.47.0 [skip ci]
### Chores
- chore: release v0.47.2
- chore: update VSCode extension package.json to v0.47.1 [skip ci]
- chore: update npm package.json to v0.47.1 [skip ci]
- chore: release v0.47.1
- chore: release v0.47.0
- chore: update VSCode extension package.json to v0.46.0 [skip ci]
- chore: update npm package.json to v0.46.0 [skip ci]
# [Version 0.47.2] - 2025-11-25
### Features
- feat: Enhance tool policy with pre-approval allowlist, improve file operation error messages, and refine tool declarations.
- feat: Integrate production-grade tool improvements system
- feat: Implement animated thinking spinner for user input submission
- feat: Add comprehensive ANSI escape sequence documentation and a new core utility module for ANSI codes.
### Bug Fixes
- fix: correct RUSTFLAGS invalid option and align system prompt with actual tool definitions
- fix: resolve all cargo clippy warnings and update rust toolchain to stable
- fix: Revise thinking spinner message to use first-person agent voice
- fix: Move thinking spinner display to after user message in transcript
- fix: Clear thinking spinner message on all agent response command types
### Refactors
- refactor: Update ThinkingSpinner struct visibility for better encapsulation
- refactor: remove redundant reasoning handling, clarify intent
### Documentation
- docs: update changelog for v0.47.1 [skip ci]
- docs: update changelog for v0.47.0 [skip ci]
### Chores
- chore: update VSCode extension package.json to v0.47.1 [skip ci]
- chore: update npm package.json to v0.47.1 [skip ci]
- chore: release v0.47.1
- chore: release v0.47.0
- chore: update VSCode extension package.json to v0.46.0 [skip ci]
- chore: update npm package.json to v0.46.0 [skip ci]
# [Version 0.47.1] - 2025-11-23
### Features
- feat: Enhance tool policy with pre-approval allowlist, improve file operation error messages, and refine tool declarations.
- feat: Integrate production-grade tool improvements system
- feat: Implement animated thinking spinner for user input submission
- feat: Add comprehensive ANSI escape sequence documentation and a new core utility module for ANSI codes.
- feat: Refactor install script, rename `run_pty_cmd` to `run_terminal_cmd`, and update installation instructions for Homebrew and NPM.
### Bug Fixes
- fix: correct RUSTFLAGS invalid option and align system prompt with actual tool definitions
- fix: resolve all cargo clippy warnings and update rust toolchain to stable
- fix: Revise thinking spinner message to use first-person agent voice
- fix: Move thinking spinner display to after user message in transcript
- fix: Clear thinking spinner message on all agent response command types
- fix: Update public re-export and documentation to reference file_helpers instead of legacy
- fix: Improve `edit_file` tool's robustness
### Refactors
- refactor: Update ThinkingSpinner struct visibility for better encapsulation
- refactor: remove redundant reasoning handling, clarify intent
- refactor: Rename `legacy` module to `file_helpers` and fix critical `edit_file` bugs related to newline handling, matching, and trailing newlines.
- refactor(mcp): Clean up unused imports
- refactor: Rename `run_terminal_cmd` to `run_pty_cmd` across documentation, examples, and tests.
### Documentation
- docs: update changelog for v0.47.0 [skip ci]
- docs: update changelog for v0.46.0 [skip ci]
- docs/mcp: integrate DEPLOYMENT_GUIDE and update INDEX navigation
- docs/mcp: Add lessons learned - project retrospective
- docs/mcp: Add team communication kit - ready-to-use materials
- docs/mcp: Add master index - 00_START_HERE.md
- docs/mcp: Add implementation guides for immediate team use
- docs/mcp: Add executive summary document
- docs/mcp: Add team announcement document
- docs: Link MCP module docs and add team guide
- docs/mcp: Add comprehensive INDEX.md for navigation
- docs/mcp: Complete documentation migration - consolidate and organize
- docs: Update and expand documentation across various topics, add a new MCP diagnostic guide, and adjust project configurations and dependencies.
### Chores
- chore: release v0.47.0
- chore: update VSCode extension package.json to v0.46.0 [skip ci]
- chore: update npm package.json to v0.46.0 [skip ci]
- chore: release v0.46.0
- chore: update VSCode extension package.json to v0.45.6 [skip ci]
- chore: update npm package.json to v0.45.6 [skip ci]
# [Version 0.47.0] - 2025-11-23
### Features
- feat: Enhance tool policy with pre-approval allowlist, improve file operation error messages, and refine tool declarations.
- feat: Integrate production-grade tool improvements system
- feat: Implement animated thinking spinner for user input submission
- feat: Add comprehensive ANSI escape sequence documentation and a new core utility module for ANSI codes.
- feat: Refactor install script, rename `run_pty_cmd` to `run_terminal_cmd`, and update installation instructions for Homebrew and NPM.
### Bug Fixes
- fix: correct RUSTFLAGS invalid option and align system prompt with actual tool definitions
- fix: resolve all cargo clippy warnings and update rust toolchain to stable
- fix: Revise thinking spinner message to use first-person agent voice
- fix: Move thinking spinner display to after user message in transcript
- fix: Clear thinking spinner message on all agent response command types
- fix: Update public re-export and documentation to reference file_helpers instead of legacy
- fix: Improve `edit_file` tool's robustness
### Refactors
- refactor: Update ThinkingSpinner struct visibility for better encapsulation
- refactor: remove redundant reasoning handling, clarify intent
- refactor: Rename `legacy` module to `file_helpers` and fix critical `edit_file` bugs related to newline handling, matching, and trailing newlines.
- refactor(mcp): Clean up unused imports
- refactor: Rename `run_terminal_cmd` to `run_pty_cmd` across documentation, examples, and tests.
### Documentation
- docs: update changelog for v0.46.0 [skip ci]
- docs/mcp: integrate DEPLOYMENT_GUIDE and update INDEX navigation
- docs/mcp: Add lessons learned - project retrospective
- docs/mcp: Add team communication kit - ready-to-use materials
- docs/mcp: Add master index - 00_START_HERE.md
- docs/mcp: Add implementation guides for immediate team use
- docs/mcp: Add executive summary document
- docs/mcp: Add team announcement document
- docs: Link MCP module docs and add team guide
- docs/mcp: Add comprehensive INDEX.md for navigation
- docs/mcp: Complete documentation migration - consolidate and organize
- docs: Update and expand documentation across various topics, add a new MCP diagnostic guide, and adjust project configurations and dependencies.
### Chores
- chore: update VSCode extension package.json to v0.46.0 [skip ci]
- chore: update npm package.json to v0.46.0 [skip ci]
- chore: release v0.46.0
- chore: update VSCode extension package.json to v0.45.6 [skip ci]
- chore: update npm package.json to v0.45.6 [skip ci]
# [Version 0.46.0] - 2025-11-21
### Features
- feat: Refactor install script, rename `run_pty_cmd` to `run_terminal_cmd`, and update installation instructions for Homebrew and NPM.
- feat: Add new tools to tool-policy and update permissions for fetch and time providers
- feat: Add default editor fallback (vi on Unix, notepad on Windows) when EDITOR/VISUAL not set
- feat: Add external editor integration with TUI suspension, alternate screen handling, and stability improvements.
- feat: update tool policies, add setup script, and enhance README with configuration details
### Bug Fixes
- fix: Update public re-export and documentation to reference file_helpers instead of legacy
- fix: Improve `edit_file` tool's robustness
- fix: Track fire-and-forget tokio::spawn tasks with JoinHandles
- fix: Apply Ratatui FAQ best practices - fix async/tokio issues
- fix: Add environment() and path() to EditorBuilder to properly detect and launch editor
- fix: Remove duplicate test block with non-existent method in zed.rs
### Refactors
- refactor: Rename `legacy` module to `file_helpers` and fix critical `edit_file` bugs related to newline handling, matching, and trailing newlines.
- refactor(mcp): Clean up unused imports
- refactor: Rename `run_terminal_cmd` to `run_pty_cmd` across documentation, examples, and tests.
- refactor: Remove static default editor, rely on try_common_editors for fallback
### Documentation
- docs/mcp: integrate DEPLOYMENT_GUIDE and update INDEX navigation
- docs/mcp: Add lessons learned - project retrospective
- docs/mcp: Add team communication kit - ready-to-use materials
- docs/mcp: Add master index - 00_START_HERE.md
- docs/mcp: Add implementation guides for immediate team use
- docs/mcp: Add executive summary document
- docs/mcp: Add team announcement document
- docs: Link MCP module docs and add team guide
- docs/mcp: Add comprehensive INDEX.md for navigation
- docs/mcp: Complete documentation migration - consolidate and organize
- docs: Update and expand documentation across various topics, add a new MCP diagnostic guide, and adjust project configurations and dependencies.
- docs: update changelog for v0.45.6 [skip ci]
- docs: add comprehensive Ratatui improvements summary
- docs: Add async improvements documentation
- docs: add Ratatui FAQ integration summary document
- docs: add Ratatui FAQ-based TUI best practices guides
- docs: Add External Editor Configuration to docs index
### Chores
- chore: update VSCode extension package.json to v0.45.6 [skip ci]
- chore: update npm package.json to v0.45.6 [skip ci]
- chore: release v0.45.6
- chore: update VSCode extension package.json to v0.45.5 [skip ci]
- chore: update npm package.json to v0.45.5 [skip ci]
# [Version 0.45.6] - 2025-11-20
### Features
- feat: Add new tools to tool-policy and update permissions for fetch and time providers
- feat: Add default editor fallback (vi on Unix, notepad on Windows) when EDITOR/VISUAL not set
- feat: Add external editor integration with TUI suspension, alternate screen handling, and stability improvements.
- feat: update tool policies, add setup script, and enhance README with configuration details
- feat: VT Code System Prompt v3 - Context Optimized Implementation
- feat: add Bash tool and remove non-existent run_pty_cmd references
- feat: implement interactive tree UI for file structure visualization
- feat: enhance diff display with full-width backgrounds and improve terminal command visibility
### Bug Fixes
- fix: Track fire-and-forget tokio::spawn tasks with JoinHandles
- fix: Apply Ratatui FAQ best practices - fix async/tokio issues
- fix: Add environment() and path() to EditorBuilder to properly detect and launch editor
- fix: Remove duplicate test block with non-existent method in zed.rs
- fix: expose shell tool to LLM by setting expose_in_llm to true
- fix: ensure development tools are always in PATH with fallback paths
- fix: remove overly complex sandbox cache clearing on PTY retry
- fix: improve loop detection for repeated tool calls
- fix: sync embedded asset for generate-agent-file.md
- fix: remove duplicate user message from conversation history
- fix: remove duplicate user message in turn loop
### Refactors
- refactor: Remove static default editor, rely on try_common_editors for fallback
- refactor: Remove sandbox functionality and streamline shell command
- refactor: eliminate wrapper layer in execute_shell_command
- refactor: use pattern matching in execute_shell_command for clarity
- refactor: simplify execute_shell_command further
- refactor: dramatically simplify execute_shell_command
- refactor: simplify execute_shell_command to skip conversion layer
- refactor: rename bash to shell and mark run_pty_cmd as deprecated
- refactor: streamline command execution error suggestions and implement unified run command executor
- refactor: streamline loop detection logic and improve non-interactive handling
### Documentation
- docs: add comprehensive Ratatui improvements summary
- docs: Add async improvements documentation
- docs: add Ratatui FAQ integration summary document
- docs: add Ratatui FAQ-based TUI best practices guides
- docs: Add External Editor Configuration to docs index
- docs: update changelog for v0.45.5 [skip ci]
- docs: Add implementation completion summary for System Prompt v3
- docs: fix misleading comment for RUN_PTY_CMD constant
- docs: add comprehensive PTY fix outcome report with complete analysis
- docs: add comprehensive PTY shell initialization fix guide
- docs: add PTY fix outcome report with validation and impact assessment
- docs: add comprehensive PTY fix summary with problem analysis and solution validation
- docs: update PTY command execution improvements documentation
- docs: add embedded assets management guide and pre-commit hook
### Chores
- chore: update VSCode extension package.json to v0.45.5 [skip ci]
- chore: update npm package.json to v0.45.5 [skip ci]
- chore: release v0.45.5
- chore: standardize default shell in workflow files and set job timeouts
- chore: update VSCode extension package.json to v0.45.4 and commit changes [skip ci]
- chore: update npm package.json to v0.45.4 [skip ci]
# [Version 0.45.5] - 2025-11-19
### Features
- feat: VT Code System Prompt v3 - Context Optimized Implementation
- feat: add Bash tool and remove non-existent run_pty_cmd references
- feat: implement interactive tree UI for file structure visualization
- feat: enhance diff display with full-width backgrounds and improve terminal command visibility
### Bug Fixes
- fix: expose shell tool to LLM by setting expose_in_llm to true
- fix: ensure development tools are always in PATH with fallback paths
- fix: remove overly complex sandbox cache clearing on PTY retry
- fix: improve loop detection for repeated tool calls
- fix: sync embedded asset for generate-agent-file.md
- fix: remove duplicate user message from conversation history
- fix: remove duplicate user message in turn loop
- fix: suppress dead_code warnings for intentionally disabled features
### Refactors
- refactor: Remove sandbox functionality and streamline shell command
- refactor: eliminate wrapper layer in execute_shell_command
- refactor: use pattern matching in execute_shell_command for clarity
- refactor: simplify execute_shell_command further
- refactor: dramatically simplify execute_shell_command
- refactor: simplify execute_shell_command to skip conversion layer
- refactor: rename bash to shell and mark run_pty_cmd as deprecated
- refactor: streamline command execution error suggestions and implement unified run command executor
- refactor: streamline loop detection logic and improve non-interactive handling
- refactor: replace dissimilar with optimized Myers diff algorithm
### Documentation
- docs: Add implementation completion summary for System Prompt v3
- docs: fix misleading comment for RUN_PTY_CMD constant
- docs: add comprehensive PTY fix outcome report with complete analysis
- docs: add comprehensive PTY shell initialization fix guide
- docs: add PTY fix outcome report with validation and impact assessment
- docs: add comprehensive PTY fix summary with problem analysis and solution validation
- docs: update PTY command execution improvements documentation
- docs: add embedded assets management guide and pre-commit hook
- docs: update changelog for v0.45.4 [skip ci]
- docs: update AGENTS.md with comprehensive agent guide and tool usage guidelines
### Chores
- chore: standardize default shell in workflow files and set job timeouts
- chore: update VSCode extension package.json to v0.45.4 and commit changes [skip ci]
- chore: update npm package.json to v0.45.4 [skip ci]
- chore: release v0.45.4
- chore: update npm package.json to v0.45.3 [skip ci]
# [Version 0.45.4] - 2025-11-17
### Bug Fixes
- fix: suppress dead_code warnings for intentionally disabled features
- fix: prevent infinite tool loops by using >= instead of >
- fix: resolve clippy warnings (range_contains, doc comments, identical blocks)
- fix: improve tool failure handling by tracking failed attempts
- fix: truncate verbose reasoning output to reduce noise during tool execution
- fix(llm): update lmstudio provider: remove stale 'For now' comment and simplify validation; update related utility and policy files
### Refactors
- refactor: replace dissimilar with optimized Myers diff algorithm
- refactor: organize documentation into docs/phases and docs/scroll subdirectories; consolidate PHASE5 and SCROLL artifacts for better maintainability
### Documentation
- docs: update AGENTS.md with comprehensive agent guide and tool usage guidelines
- docs: update changelog for v0.45.3 [skip ci]
- docs: reorganize root-level docs into docs/ subdirectories per AGENTS.md
- docs: clarify run_pty_cmd usage for git, cargo, and one-off shell commands
### Style Changes
- style: apply cargo fmt
### Chores
- chore: update npm package.json to v0.45.3 [skip ci]
- chore: release v0.45.3
- chore: update npm package.json to v0.45.2 [skip ci]
# [Version 0.45.3] - 2025-11-17
### Features
- feat: implement token-based truncation for tool outputs and update configuration
### Bug Fixes
- fix: prevent infinite tool loops by using >= instead of >
- fix: resolve clippy warnings (range_contains, doc comments, identical blocks)
- fix: improve tool failure handling by tracking failed attempts
- fix: truncate verbose reasoning output to reduce noise during tool execution
- fix(llm): update lmstudio provider: remove stale 'For now' comment and simplify validation; update related utility and policy files
### Refactors
- refactor: organize documentation into docs/phases and docs/scroll subdirectories; consolidate PHASE5 and SCROLL artifacts for better maintainability
### Documentation
- docs: reorganize root-level docs into docs/ subdirectories per AGENTS.md
- docs: clarify run_pty_cmd usage for git, cargo, and one-off shell commands
- docs: update changelog for v0.45.2 [skip ci]
- docs: update changelog for v0.45.1 [skip ci]
### Style Changes
- style: apply cargo fmt
### Chores
- chore: update npm package.json to v0.45.2 [skip ci]
- chore: release v0.45.2
- chore: release v0.45.1
- chore: update npm package.json to v0.45.0 [skip ci]
# [Version 0.45.2] - 2025-11-17
### Features
- feat: implement token-based truncation for tool outputs and update configuration
### Refactors
- refactor(runloop): extract tool pipeline into and add — reduce run loop complexity
### Documentation
- docs: update changelog for v0.45.1 [skip ci]
- docs: update changelog for v0.45.0 [skip ci]
### Chores
- chore: release v0.45.1
- chore: update npm package.json to v0.45.0 [skip ci]
- chore: release v0.45.0
- chore: update GitHub Actions workflows for improved performance and consistency; adjust dependency management and environment variables
- chore(runloop): make session.rs minimal exposing slash_commands
- chore(runloop): remove session.rs contents to extract run loop
- chore(runloop): Extract run_single_agent_loop_unified to run_loop.rs
- chore: update npm package.json to v0.44.1 [skip ci]
# [Version 0.45.1] - 2025-11-17
### Features
- feat: implement token-based truncation for tool outputs and update configuration
### Refactors
- refactor(runloop): extract tool pipeline into and add — reduce run loop complexity
### Documentation
- docs: update changelog for v0.45.0 [skip ci]
### Chores
- chore: update npm package.json to v0.45.0 [skip ci]
- chore: release v0.45.0
- chore: update GitHub Actions workflows for improved performance and consistency; adjust dependency management and environment variables
- chore(runloop): make session.rs minimal exposing slash_commands
- chore(runloop): remove session.rs contents to extract run loop
- chore(runloop): Extract run_single_agent_loop_unified to run_loop.rs
- chore: update npm package.json to v0.44.1 [skip ci]
# [Version 0.45.0] - 2025-11-16
### Bug Fixes
- fix: update Claude model identifiers and descriptions for accuracy
### Refactors
- refactor(runloop): extract tool pipeline into and add — reduce run loop complexity
### Documentation
- docs: update changelog for v0.44.1 [skip ci]
- docs: update changelog for v0.44.0 [skip ci]
### Chores
- chore: update GitHub Actions workflows for improved performance and consistency; adjust dependency management and environment variables
- chore(runloop): make session.rs minimal exposing slash_commands
- chore(runloop): remove session.rs contents to extract run loop
- chore(runloop): Extract run_single_agent_loop_unified to run_loop.rs
- chore: update npm package.json to v0.44.1 [skip ci]
- chore: release v0.44.1
- chore: release v0.44.0
- chore: update configuration files for VT Code support
- chore: update npm package.json to v0.43.17 [skip ci]
# [Version 0.44.1] - 2025-11-15
### Bug Fixes
- fix: update Claude model identifiers and descriptions for accuracy
### Documentation
- docs: update changelog for v0.44.0 [skip ci]
- docs: update changelog for v0.43.17 [skip ci]
- docs: update changelog for v0.43.16 [skip ci]
### Chores
- chore: release v0.44.0
- chore: update configuration files for VT Code support
- chore: update npm package.json to v0.43.17 [skip ci]
- chore: release v0.43.17
- chore: release v0.43.16
- chore: update npm package.json to v0.43.15 [skip ci]
# [Version 0.44.0] - 2025-11-15
### Documentation
- docs: update changelog for v0.43.17 [skip ci]
- docs: update changelog for v0.43.16 [skip ci]
### Chores
- chore: update configuration files for VT Code support
- chore: update npm package.json to v0.43.17 [skip ci]
- chore: release v0.43.17
- chore: release v0.43.16
- chore: update npm package.json to v0.43.15 [skip ci]
# [Version 0.43.17] - 2025-11-15
### Documentation
- docs: update changelog for v0.43.16 [skip ci]
- docs: update changelog for v0.43.15 [skip ci]
- docs: update changelog for v0.43.14 [skip ci]
- docs: update changelog for v0.43.13 [skip ci]
- docs: update changelog for v0.43.12 [skip ci]
### Chores
- chore: release v0.43.16
- chore: update npm package.json to v0.43.15 [skip ci]
- chore: release v0.43.15
- chore: release v0.43.14
- chore: release v0.43.13
- chore: release v0.43.12
- chore: update npm package.json to v0.43.11 [skip ci]
### Features
- feat(openai): add `prompt_cache_retention` option in vtcode.toml to control Responses API cache retention (e.g., "24h")
# [Version 0.43.16] - 2025-11-15
### Documentation
- docs: update changelog for v0.43.15 [skip ci]
- docs: update changelog for v0.43.14 [skip ci]
- docs: update changelog for v0.43.13 [skip ci]
- docs: update changelog for v0.43.12 [skip ci]
### Chores
- chore: update npm package.json to v0.43.15 [skip ci]
- chore: release v0.43.15
- chore: release v0.43.14
- chore: release v0.43.13
- chore: release v0.43.12
- chore: update npm package.json to v0.43.11 [skip ci]
# [Version 0.43.15] - 2025-11-14
### Documentation
- docs: update changelog for v0.43.14 [skip ci]
- docs: update changelog for v0.43.13 [skip ci]
- docs: update changelog for v0.43.12 [skip ci]
- docs: update changelog for v0.43.11 [skip ci]
### Chores
- chore: release v0.43.14
- chore: release v0.43.13
- chore: release v0.43.12
- chore: update npm package.json to v0.43.11 [skip ci]
- chore: release v0.43.11
- chore: update npm package.json to v0.43.10 [skip ci]
# [Version 0.43.14] - 2025-11-14
### Documentation
- docs: update changelog for v0.43.13 [skip ci]
- docs: update changelog for v0.43.12 [skip ci]
- docs: update changelog for v0.43.11 [skip ci]
### Chores
- chore: release v0.43.13
- chore: release v0.43.12
- chore: update npm package.json to v0.43.11 [skip ci]
- chore: release v0.43.11
- chore: update npm package.json to v0.43.10 [skip ci]
# [Version 0.43.13] - 2025-11-14
### Documentation
- docs: update changelog for v0.43.12 [skip ci]
- docs: update changelog for v0.43.11 [skip ci]
### Chores
- chore: release v0.43.12
- chore: update npm package.json to v0.43.11 [skip ci]
- chore: release v0.43.11
- chore: update npm package.json to v0.43.10 [skip ci]
# [Version 0.43.12] - 2025-11-14
### Documentation
- docs: update changelog for v0.43.11 [skip ci]
### Chores
- chore: update npm package.json to v0.43.11 [skip ci]
- chore: release v0.43.11
- chore: update npm package.json to v0.43.10 [skip ci]
# [Version 0.43.11] - 2025-11-13
### Documentation
- docs: update changelog for v0.43.10 [skip ci]
- docs: update changelog for v0.43.9 [skip ci]
### Chores
- chore: update npm package.json to v0.43.10 [skip ci]
- chore: release v0.43.10
- chore: release v0.43.9
- chore: update npm package.json to v0.43.8 [skip ci]
# [Version 0.43.10] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
### Documentation
- docs: update changelog for v0.43.9 [skip ci]
- docs: update changelog for v0.43.8 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
### Chores
- chore: release v0.43.9
- chore: update npm package.json to v0.43.8 [skip ci]
- chore: release v0.43.8
- chore: release v0.43.7
- chore: update npm package.json to v0.43.6 [skip ci]
# [Version 0.43.9] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
### Documentation
- docs: update changelog for v0.43.8 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
### Chores
- chore: update npm package.json to v0.43.8 [skip ci]
- chore: release v0.43.8
- chore: release v0.43.7
- chore: update npm package.json to v0.43.6 [skip ci]
# [Version 0.43.8] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
- feat(web_fetch): Introduce Web Fetch tool with security configurations
- feat: Implement token-based truncation for tool output rendering
- feat: Enhance command execution with additional PATH entries and environment variable handling
### Refactors
- refactor: replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization
- refactor: update InlineTextStyle to include bg_color and effects in snapshot tests
- refactor: update command execution in tests and remove unused imports
### Documentation
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.6 [skip ci]
- docs: Add truncation audit and remove unused terminal output line-limit constants
### Chores
- chore: release v0.43.7
- chore: update npm package.json to v0.43.6 [skip ci]
- chore: release v0.43.6
- chore: update npm package.json to v0.43.5 [skip ci]
# [Version 0.43.7] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
- feat(web_fetch): Introduce Web Fetch tool with security configurations
- feat: Implement token-based truncation for tool output rendering
- feat: Enhance command execution with additional PATH entries and environment variable handling
### Refactors
- refactor: replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization
- refactor: update InlineTextStyle to include bg_color and effects in snapshot tests
- refactor: update command execution in tests and remove unused imports
### Documentation
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.6 [skip ci]
- docs: Add truncation audit and remove unused terminal output line-limit constants
### Chores
- chore: update npm package.json to v0.43.6 [skip ci]
- chore: release v0.43.6
- chore: update npm package.json to v0.43.5 [skip ci]
# [Version 0.43.7] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
- feat(web_fetch): Introduce Web Fetch tool with security configurations
- feat: Implement token-based truncation for tool output rendering
- feat: Enhance command execution with additional PATH entries and environment variable handling
### Refactors
- refactor: replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization
- refactor: update InlineTextStyle to include bg_color and effects in snapshot tests
- refactor: update command execution in tests and remove unused imports
### Documentation
- docs: update changelog for v0.43.7 [skip ci]
- docs: update changelog for v0.43.6 [skip ci]
- docs: Add truncation audit and remove unused terminal output line-limit constants
### Chores
- chore: update npm package.json to v0.43.6 [skip ci]
- chore: release v0.43.6
- chore: update npm package.json to v0.43.5 [skip ci]
# [Version 0.43.7] - 2025-11-13
### Features
- feat(ripgrep): Add automatic installation and management for ripgrep dependency
- feat(loop_detection): Refactor loop hang detection for improved accuracy and user experience
- feat(web_fetch): Introduce Web Fetch tool with security configurations
- feat: Implement token-based truncation for tool output rendering
- feat: Enhance command execution with additional PATH entries and environment variable handling
### Refactors
- refactor: replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization
- refactor: update InlineTextStyle to include bg_color and effects in snapshot tests
- refactor: update command execution in tests and remove unused imports
### Documentation
- docs: update changelog for v0.43.6 [skip ci]
- docs: Add truncation audit and remove unused terminal output line-limit constants
### Chores
- chore: update npm package.json to v0.43.6 [skip ci]
- chore: release v0.43.6
- chore: update npm package.json to v0.43.5 [skip ci]
# [Version 0.43.6] - 2025-11-12
### Features
- feat(web_fetch): Introduce Web Fetch tool with security configurations
- feat: Implement token-based truncation for tool output rendering
- feat: Enhance command execution with additional PATH entries and environment variable handling
### Refactors
- refactor: replace cargo_bin_cmd with assert_cmd in CLI tests and simplify InlineTextStyle initialization
- refactor: update InlineTextStyle to include bg_color and effects in snapshot tests
- refactor: update command execution in tests and remove unused imports
### Documentation
- docs: Add truncation audit and remove unused terminal output line-limit constants
- docs: update changelog for v0.43.5 [skip ci]
### Chores
- chore: update npm package.json to v0.43.5 [skip ci]
- chore: release v0.43.5
- chore: update npm package.json to v0.43.4 [skip ci]
# [Version 0.43.5] - 2025-11-11
### Features
- feat: Implement permission system with command resolution, audit logging, and caching
### Improvements
- improve: Enhanced token approximation algorithm with median-based heuristics for fallback tokenization
- improve: Fixed token counting fallback to use consistent 3.5 chars/token ratio across head/tail sections
- improve: Eliminated async token counting overhead by using fast character-based fallback estimation
- improve: Optimized tail content collection from O(n²) string operations to O(n) with Vec collection
- improve: Added String pre-allocation with capacity to reduce memory allocations during truncation
- improve: Improved median-based token estimation to handle edge cases (zero word count, whitespace-heavy content)
- improve: Optimized result assembly with in-place string building and size pre-calculation
- improve: Increased code fence block display limit from 200 → 500 lines with better truncation messaging
- improve: Increased diff preview display limit from 300 → 500 lines with improved user guidance
- improve: Added comprehensive module-level documentation for token-aware truncation strategy
- improve: Clarified token budget messaging to users about what content is preserved
### Refactors
- refactor: Remove unused audit log and history navigation methods
- refactor: Remove references to ast_grep_search from documentation and tool policies
- refactor: Phase 2 Step 4 - migrate remaining input methods and word navigation
- refactor: Phase 2 Step 3 - migrate clear_input() and reset_history_navigation()
- refactor: Phase 2 Step 2 - add manager sync helper methods
- refactor: Phase 2 Step 1 - add manager fields to Session struct
- refactor: extract input history navigation logic fix in InputManager
### Documentation
- docs: update changelog for v0.43.4 [skip ci]
- docs: update Phase 2 progress - Step 4 complete with all input methods migrated
- docs: add VT Code execution policy documentation and update command validation
### Chores
- chore: update npm package.json to v0.43.4 [skip ci]
- chore: release v0.43.4
- chore: update npm package.json to v0.43.3 [skip ci]
# [Version 0.43.4] - 2025-11-10
### Features
- feat: Implement permission system with command resolution, audit logging, and caching
- feat: Add Git color configuration support and theme management
- feat: Add Styling Quick Start Guide and Refactor Completion Report
- feat: add theme_parser module for Git/LS_COLORS configuration parsing
- feat: complete phase 1 anstyle integration - effects and background colors
- feat: Integrate anstyle-parse for ANSI escape sequence handling
- feat: implement styling refactor - centralize color palettes and style helpers
### Bug Fixes
- fix: redirect logging to stderr to prevent stdout pollution in install script
### Refactors
- refactor: Remove unused audit log and history navigation methods
- refactor: Remove references to ast_grep_search from documentation and tool policies
- refactor: Phase 2 Step 4 - migrate remaining input methods and word navigation
- refactor: Phase 2 Step 3 - migrate clear_input() and reset_history_navigation()
- refactor: Phase 2 Step 2 - add manager sync helper methods
- refactor: Phase 2 Step 1 - add manager fields to Session struct
- refactor: extract input history navigation logic fix in InputManager
- refactor(styling): implement central style helpers and diff color palette
- refactor: improve styling consistency with bold_color() and ColorPalette
- refactor: implement styling suggestions from STYLING_REFACTOR_GUIDE
- refactor: implement styling refactor from guide - centralize color/style management
### Documentation
- docs: update Phase 2 progress - Step 4 complete with all input methods migrated
- docs: add VT Code execution policy documentation and update command validation
- docs: update changelog for v0.43.3 [skip ci]
- docs: add comprehensive styling documentation index
- docs: add Phase 2 planning and implementation guides for advanced styling features
- docs: add session summary for phase 1 styling integration completion
- docs: add phase 1 completion summary - all criteria met
- docs: add styling implementation completion status
- docs: add styling implementation completion status
- docs: update installation guides with CDN caching troubleshooting and fix details
### Chores
- chore: update npm package.json to v0.43.3 [skip ci]
- chore: release v0.43.3
- chore: update install script to log messages to stderr and bump version to 0.43.2
- chore: update npm package.json to v0.43.2 [skip ci]
# [Version 0.43.3] - 2025-11-09
### Features
- feat: Add Git color configuration support and theme management
- feat: Add Styling Quick Start Guide and Refactor Completion Report
- feat: add theme_parser module for Git/LS_COLORS configuration parsing
- feat: complete phase 1 anstyle integration - effects and background colors
- feat: Integrate anstyle-parse for ANSI escape sequence handling
- feat: implement styling refactor - centralize color palettes and style helpers
### Bug Fixes
- fix: redirect logging to stderr to prevent stdout pollution in install script
- fix: optimize list_files tool for improved pagination and reduce default page size
- fix: update LLM provider and models to use Ollama
- fix: revert extension.toml to valid Zed format
### Refactors
- refactor(styling): implement central style helpers and diff color palette
- refactor: improve styling consistency with bold_color() and ColorPalette
- refactor: implement styling suggestions from STYLING_REFACTOR_GUIDE
- refactor: implement styling refactor from guide - centralize color/style management
- refactor: integrate CommandBuilder throughout commands module
- refactor: restructure zed-extension to modular architecture with comprehensive error handling and caching
### Documentation
- docs: add comprehensive styling documentation index
- docs: add Phase 2 planning and implementation guides for advanced styling features
- docs: add session summary for phase 1 styling integration completion
- docs: add phase 1 completion summary - all criteria met
- docs: add styling implementation completion status
- docs: add styling implementation completion status
- docs: update installation guides with CDN caching troubleshooting and fix details
- docs: update changelog for v0.43.2 [skip ci]
- docs: add file listing output behavior pattern to AGENTS.md
- docs: add comprehensive final improvements summary
- docs: update STATUS with improvements session results
### Chores
- chore: update install script to log messages to stderr and bump version to 0.43.2
- chore: update npm package.json to v0.43.2 [skip ci]
- chore: release v0.43.2
- chore: update npm package.json to v0.43.1 [skip ci]
# [Version 0.43.2] - 2025-11-09
### Bug Fixes
- fix: optimize list_files tool for improved pagination and reduce default page size
- fix: update LLM provider and models to use Ollama
- fix: revert extension.toml to valid Zed format
### Refactors
- refactor: integrate CommandBuilder throughout commands module
- refactor: restructure zed-extension to modular architecture with comprehensive error handling and caching
### Documentation
- docs: add file listing output behavior pattern to AGENTS.md
- docs: add comprehensive final improvements summary
- docs: update STATUS with improvements session results
- docs: update changelog for v0.43.1 [skip ci]
- docs: Add release readiness confirmation document
- docs: Add release action checklist for v0.43.0
- docs: Add comprehensive v0.43.0 release summary
### Chores
- chore: update npm package.json to v0.43.1 [skip ci]
- chore: release v0.43.1
# [Version 0.43.1] - 2025-11-09
### Features
- feat: Implement Agent Communication Protocol (ACP) integration
### Documentation
- docs: Add release readiness confirmation document
- docs: Add release action checklist for v0.43.0
- docs: Add comprehensive v0.43.0 release summary
- docs: Update ACP implementation summary and usage patterns
- docs: Add ACP next steps and release checklist
- docs: Add ACP implementation completion summary
### Chores
- chore: release v0.43.0
- chore: bump version to 0.43.0 for ACP release
- chore: update npm package.json to v0.42.20 [skip ci]
### Features
- feat: Implement Agent Communication Protocol (ACP) integration for multi-agent orchestration
- feat: Add ACP client with sync/async RPC methods
- feat: Implement agent discovery and registry system
- feat: Add type-safe message protocol with correlation ID tracking
- feat: Create MCP tools: acp_call, acp_discover, acp_health for agent communication
- feat: Integrate ACP with Zed editor for terminal command execution
- feat: Support distributed agent workflows via HTTP-based RPC
### Documentation
- docs: Add comprehensive ACP integration guide
- docs: Add ACP quick reference for developers
- docs: Add ACP client API documentation and examples
- docs: Add implementation completion summary
- docs: Add release checklist and next steps guide
### Testing
- test: Add full test coverage for ACP client (6 unit tests)
- test: Add ACP tool integration tests
- test: Add distributed workflow example
## [Version 0.42.20] - 2025-11-09
### Features
- feat: Implement tool approval dialog with enhanced UX and risk assessment
- feat: Step 8 - Implement tool versioning and compatibility checking
- feat: Step 7 - Observability & Metrics system for MCP execution
- feat: implement all 5 MCP code execution steps from Anthropic recommendations
- feat: Step 2 Phase 2 - IPC handler integration for tool invocation
- feat: Step 2 - Code executor with SDK generation and IPC
### Bug Fixes
- fix: Resolve compilation warnings and duplicate test module
- fix: remove unused import in code_executor
### Documentation
- docs: update changelog for v0.42.19 [skip ci]
- docs: update changelog for v0.42.18 [skip ci]
- docs: add comprehensive tool configuration status document
- docs: add agent prompt optimization summary
- docs: Add MCP quick reference guide for fast lookup
- docs: Add comprehensive MCP implementation status report
- docs: Complete 9-step MCP code execution roadmap with Steps 8-9 designs
- docs: Add Step 6 integration testing guide and test scenarios
- docs: Update Step 2 completion status and add SDK examples
### Chores
- chore: update npm package.json to v0.42.19 [skip ci]
- chore: release v0.42.19
- chore: release v0.42.18
- chore: finalize tool configuration and system prompt updates
- chore: update npm package.json to v0.42.17 [skip ci]
# [Version 0.42.19] - 2025-11-08
### Features
- feat: Step 8 - Implement tool versioning and compatibility checking
- feat: Step 7 - Observability & Metrics system for MCP execution
- feat: implement all 5 MCP code execution steps from Anthropic recommendations
- feat: Step 2 Phase 2 - IPC handler integration for tool invocation
- feat: Step 2 - Code executor with SDK generation and IPC
- feat: Add comprehensive timeout implementation summary and configuration details
- feat: Implement configurable MCP initialization and tool execution timeouts
- feat: Add OpenRouter Interleaved Thinking Implementation Plan and Quick Reference
### Bug Fixes
- fix: Resolve compilation warnings and duplicate test module
- fix: remove unused import in code_executor
### Documentation
- docs: update changelog for v0.42.18 [skip ci]
- docs: add comprehensive tool configuration status document
- docs: add agent prompt optimization summary
- docs: Add MCP quick reference guide for fast lookup
- docs: Add comprehensive MCP implementation status report
- docs: Complete 9-step MCP code execution roadmap with Steps 8-9 designs
- docs: Add Step 6 integration testing guide and test scenarios
- docs: Update Step 2 completion status and add SDK examples
- docs: update changelog for v0.42.17 [skip ci]
### Chores
- chore: release v0.42.18
- chore: finalize tool configuration and system prompt updates
- chore: update npm package.json to v0.42.17 [skip ci]
- chore: release v0.42.17
- chore: update documentation and code structure for clarity
- chore: update npm package.json to v0.42.16 [skip ci]
# [Version 0.42.18] - 2025-11-08
### Features
- feat: Step 8 - Implement tool versioning and compatibility checking
- feat: Step 7 - Observability & Metrics system for MCP execution
- feat: implement all 5 MCP code execution steps from Anthropic recommendations
- feat: Step 2 Phase 2 - IPC handler integration for tool invocation
- feat: Step 2 - Code executor with SDK generation and IPC
- feat: Add comprehensive timeout implementation summary and configuration details
- feat: Implement configurable MCP initialization and tool execution timeouts
- feat: Add OpenRouter Interleaved Thinking Implementation Plan and Quick Reference
### Bug Fixes
- fix: Resolve compilation warnings and duplicate test module
- fix: remove unused import in code_executor
### Documentation
- docs: add comprehensive tool configuration status document
- docs: add agent prompt optimization summary
- docs: Add MCP quick reference guide for fast lookup
- docs: Add comprehensive MCP implementation status report
- docs: Complete 9-step MCP code execution roadmap with Steps 8-9 designs
- docs: Add Step 6 integration testing guide and test scenarios
- docs: Update Step 2 completion status and add SDK examples
- docs: update changelog for v0.42.17 [skip ci]
### Chores
- chore: finalize tool configuration and system prompt updates
- chore: update npm package.json to v0.42.17 [skip ci]
- chore: release v0.42.17
- chore: update documentation and code structure for clarity
- chore: update npm package.json to v0.42.16 [skip ci]
# [Version 0.42.17] - 2025-11-08
### Features
- feat: Add comprehensive timeout implementation summary and configuration details
- feat: Implement configurable MCP initialization and tool execution timeouts
- feat: Add OpenRouter Interleaved Thinking Implementation Plan and Quick Reference
### Bug Fixes
- fix: update tool policies and disable time provider in configuration
### Documentation
- docs: update changelog for v0.42.16 [skip ci]
### Chores
- chore: update documentation and code structure for clarity
- chore: update npm package.json to v0.42.16 [skip ci]
- chore: release v0.42.16
- chore: update npm package.json to v0.42.15 [skip ci]
# [Version 0.42.16] - 2025-11-08
### Bug Fixes
- fix: update tool policies and disable time provider in configuration
### Documentation
- docs: update changelog for v0.42.15 [skip ci]
### Chores
- chore: update npm package.json to v0.42.15 [skip ci]
- chore: release v0.42.15
- chore: update package name and publishing instructions for npmjs.com and GitHub Packages
- chore: update npm package.json to v0.42.14 [skip ci]
# [Version 0.42.15] - 2025-11-08
### Bug Fixes
- fix: revert version in package.json to 0.42.13
### Documentation
- docs: update changelog for v0.42.14 [skip ci]
### Chores
- chore: update package name and publishing instructions for npmjs.com and GitHub Packages
- chore: update npm package.json to v0.42.14 [skip ci]
- chore: release v0.42.14
# [Version 0.42.14] - 2025-11-08
### Features
- feat: Remove deprecated tool and add test_tool to policy
- feat: Implement NPM package publishing for VT Code
- feat: add configurable LLM generation parameters in vtcode.toml
### Bug Fixes
- fix: revert version in package.json to 0.42.13
### Documentation
- docs: update changelog for v0.42.13 [skip ci]
### Chores
- chore: release v0.42.13
# [Version 0.42.13] - 2025-11-08
### Features
- feat: Remove deprecated tool and add test_tool to policy
- feat: Implement NPM package publishing for VT Code
- feat: add configurable LLM generation parameters in vtcode.toml
### Bug Fixes
- fix: redirect print functions to stderr to avoid command substitution issues
- fix: use temporary file approach with awk for changelog updates on macOS
- fix: use perl instead of awk for changelog updates on macOS
- fix: use awk instead of sed for changelog updates on macOS
- fix: escape newlines properly in sed command for macOS
### Refactors
- refactor: update tool policies and improve MCP tool handling
### Documentation
- docs: update changelog for v0.42.12 [skip ci]
### Chores
- chore: release v0.42.12
- chore: release vscode extension v0.42.18
- chore: release vscode extension v0.42.17
- chore: release vscode extension v0.42.16
# [Version 0.42.12] - 2025-11-08
### Features
- feat: update vtcode.toml configuration for new model provider
### Bug Fixes
- fix: redirect print functions to stderr to avoid command substitution issues
- fix: use temporary file approach with awk for changelog updates on macOS
- fix: use perl instead of awk for changelog updates on macOS
- fix: use awk instead of sed for changelog updates on macOS
- fix: escape newlines properly in sed command for macOS
- fix: update Moonshot model references from KIMI_K2_THINKING_HEAVY to KIMI_K2_THINKING_TURBO
### Refactors
- refactor: update tool policies and improve MCP tool handling
### Documentation
- docs: update changelog for v0.42.11 [skip ci]
- docs: update changelog for v0.42.10 [skip ci]
### Chores
- chore: release vscode extension v0.42.18
- chore: release vscode extension v0.42.17
- chore: release vscode extension v0.42.16
- chore: release v0.42.11
- chore: release v0.42.10
# [Version 0.42.11] - 2025-11-07
### Features
- feat: update vtcode.toml configuration for new model provider
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: update Moonshot model references from KIMI_K2_THINKING_HEAVY to KIMI_K2_THINKING_TURBO
- fix: add Debug trait to MessageStyle enum
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Refactors
- refactor: remove unused app constant and update elicitation capability handling
### Documentation
- docs: update changelog for v0.42.10 [skip ci]
- docs: update changelog for v0.42.9 [skip ci]
- docs: update changelog for v0.42.8 [skip ci]
- docs: update changelog for v0.42.7 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
### Chores
- chore: release v0.42.10
- chore: release v0.42.9
- chore: release v0.42.8
- chore: release v0.42.7
- chore: release v0.42.6
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
# [Version 0.42.10] - 2025-11-07
### Features
- feat: update vtcode.toml configuration for new model provider
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: add Debug trait to MessageStyle enum
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Refactors
- refactor: remove unused app constant and update elicitation capability handling
### Documentation
- docs: update changelog for v0.42.9 [skip ci]
- docs: update changelog for v0.42.8 [skip ci]
- docs: update changelog for v0.42.7 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
### Chores
- chore: release v0.42.9
- chore: release v0.42.8
- chore: release v0.42.7
- chore: release v0.42.6
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
# [Version 0.42.9] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: add Debug trait to MessageStyle enum
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Refactors
- refactor: remove unused app constant and update elicitation capability handling
### Documentation
- docs: update changelog for v0.42.8 [skip ci]
- docs: update changelog for v0.42.7 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.8
- chore: release v0.42.7
- chore: release v0.42.6
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.8] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: add Debug trait to MessageStyle enum
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.7 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.7
- chore: release v0.42.6
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.7] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.6
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.6] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: remove jsonschema dependency from mcp-types in Cargo.lock
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.6] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: remove mcp-types configuration from release.toml
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.6 [skip ci]
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.6] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: ensure publish is set to false for mcp-types in release.toml
- fix: update mcp-types version to 0.1.1 in Cargo.lock
- fix: update mcp-types version to 0.1.1 in Cargo.toml
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.5 [skip ci]
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.5
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.5] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: ensure publish is set to false in Cargo.toml
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.4 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.4
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.4] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.3
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.3] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: remove .cargo_vcs_info.json and update vtcode-core dependency version in Cargo.toml
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.3 [skip ci]
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.3] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Bug Fixes
- fix: update mcp-types dependency path in Cargo.toml and add jsonschema to dependencies
### Documentation
- docs: update changelog for v0.42.2 [skip ci]
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.2
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.2] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Documentation
- docs: update changelog for v0.42.1 [skip ci]
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.1
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.1] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Documentation
- docs: update changelog for v0.43.0 [skip ci]
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.43.0] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Documentation
- docs: update changelog for v0.42.0 [skip ci]
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.42.0
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.42.0] - 2025-11-07
### Features
- feat: add Kimi K2 Thinking model support and update Moonshot provider logic
### Documentation
- docs: update changelog for v0.41.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.41.0
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.41.0] - 2025-11-07
### Documentation
- docs: update changelog for v0.40.1 [skip ci]
- docs: update changelog for v0.40.0 [skip ci]
- docs: update changelog for v0.40.1 [skip ci]
### Chores
- chore: release v0.40.1
- chore: release v0.40.0
- chore: release v0.40.1
# [Version 0.40.1] - 2025-11-06
### Features
- feat: Enhance workspace trust and automation features
- feat: add workspace trust request functionality
- feat: Add VT Code Chat extension with MCP integration
- feat: add experimental smart summarization feature for conversation compression
- feat: replace tempfile with assert_fs for improved temporary directory handling
### Bug Fixes
- fix: add wasm32-wasip2 component to toolchain configuration
### Refactors
- refactor: remove wasm32-wasip2 component from toolchain configuration and clean up test imports
- refactor: clean up test module by removing unused imports and structures
- refactor: replace assert_fs::prelude with tempfile::tempdir in tests
- refactor: remove unused zed-extension files and grammars
- refactor: update tool policies, exclude zed-extension from workspace, and upgrade zed_extension_api dependency
- refactor: update tool policy and improve command handling; streamline error messages and enhance telemetry logging
- refactor: update extension ID and name for consistency
- refactor: remove unused imports and streamline timeout error handling
### Documentation
- docs: update changelog for v0.40.0 [skip ci]
- docs: add troubleshooting section for development installation
### Chores
- chore: release v0.40.0
# [Version 0.40.0] - 2025-11-06
### Features
- feat: Enhance workspace trust and automation features
- feat: add workspace trust request functionality
- feat: Add VT Code Chat extension with MCP integration
- feat: add experimental smart summarization feature for conversation compression
- feat: replace tempfile with assert_fs for improved temporary directory handling
### Bug Fixes
- fix: add wasm32-wasip2 component to toolchain configuration
### Refactors
- refactor: remove wasm32-wasip2 component from toolchain configuration and clean up test imports
- refactor: clean up test module by removing unused imports and structures
- refactor: replace assert_fs::prelude with tempfile::tempdir in tests
- refactor: remove unused zed-extension files and grammars
- refactor: update tool policies, exclude zed-extension from workspace, and upgrade zed_extension_api dependency
- refactor: update tool policy and improve command handling; streamline error messages and enhance telemetry logging
- refactor: update extension ID and name for consistency
- refactor: remove unused imports and streamline timeout error handling
### Documentation
- docs: add troubleshooting section for development installation
- docs: update changelog for v0.39.13 [skip ci]
### Chores
- chore: release v0.39.13
# [Version 0.39.13] - 2025-11-03
### Features
- feat: disable Docker usage by default in build script and update Cross.toml comments
### Documentation
- docs: update changelog for v0.39.12 [skip ci]
### Chores
- chore: release v0.39.12
# [Version 0.39.12] - 2025-11-03
### Features
- feat: disable Docker usage by default in build script and update Cross.toml comments
- feat: add initial implementation of VT Code Zed extension with icons, themes, and logging commands
### Documentation
- docs: update changelog for v0.39.11 [skip ci]
### Chores
- chore: release v0.39.11
- chore: update dependencies, enhance README, and add diagnostics commands for Zed extension
# [Version 0.39.11] - 2025-11-03
### Features
- feat: add initial implementation of VT Code Zed extension with icons, themes, and logging commands
### Refactors
- refactor: remove npm package support and update installation instructions
- refactor: update docs.rs URL and improve response handling in release script
### Documentation
- docs: update changelog for v0.39.10 [skip ci]
- docs: update changelog for v0.39.9 [skip ci]
### Chores
- chore: update dependencies, enhance README, and add diagnostics commands for Zed extension
- chore: release v0.39.10
- chore: release v0.39.9
# [Version 0.39.10] - 2025-11-03
### Refactors
- refactor: remove npm package support and update installation instructions
- refactor: update docs.rs URL and improve response handling in release script
### Documentation
- docs: update changelog for v0.39.9 [skip ci]
- docs: update changelog for v0.39.8 [skip ci]
### Chores
- chore: release v0.39.9
- chore: release v0.39.8
# [Version 0.39.9] - 2025-11-03
### Refactors
- refactor: remove npm package support and update installation instructions
- refactor: update docs.rs URL and improve response handling in release script
### Documentation
- docs: update changelog for v0.39.8 [skip ci]
### Chores
- chore: release v0.39.8
# [Version 0.39.8] - 2025-11-03
### Bug Fixes
- fix: restore npm/package.json file removed in error
### Documentation
- docs: update changelog for v0.39.7 [skip ci]
### Chores
- chore: release v0.39.7
- chore: update npm package to v0.39.7
- chore: update Cross.toml and release script for improved environment variable handling
# [Version 0.39.7] - 2025-11-03
### Bug Fixes
- fix: restore npm/package.json file removed in error
### Documentation
- docs: update changelog for v0.39.6 [skip ci]
### Chores
- chore: update npm package to v0.39.7
- chore: update Cross.toml and release script for improved environment variable handling
- chore: release v0.39.6
- chore: update npm package to v0.39.6
# [Version 0.39.6] - 2025-11-03
### Features
- feat: add cross-compilation configuration and documentation
- feat: Enhance development and release process for VT Code extension
### Bug Fixes
- fix: update changelog generation to handle date formatting correctly
- fix: rename VT Code Update Plan tool for consistency
- fix: update language model tool properties for VT Code Update Plan
### Refactors
- refactor: remove unused IdeContextBridge and clean up session initialization
### Documentation
- docs: update changelog for v0.39.5 [skip ci]
- docs: update changelog for v0.39.4 [skip ci]
- docs: update changelog for v0.39.3 [skip ci]
- docs: update homebrew installation to use core tap
### Chores
- chore: update npm package to v0.39.6
- chore: release v0.39.5
- chore: update npm package to v0.39.5
- chore: release v0.39.4
- chore: update npm package to v0.39.4
- chore: release v0.39.3
- chore: update npm package to v0.39.3
- chore: update version to 0.1.1 and add release date to changelog
- chore(deps): bump the cargo-monthly-rollup group across 1 directory with 28 updates
### Features
- feat: add cross-compilation configuration and documentation|
- feat: Enhance development and release process for VT Code extension|
### Bug Fixes
- fix: update changelog generation to handle date formatting correctly|
- fix: rename VT Code Update Plan tool for consistency|
- fix: update language model tool properties for VT Code Update Plan|
### Refactors
- refactor: remove unused IdeContextBridge and clean up session initialization|
### Documentation
- docs: update changelog for v0.39.4 [skip ci]|
- docs: update changelog for v0.39.3 [skip ci]|
- docs: update homebrew installation to use core tap|
- docs: update changelog for v0.39.2 [skip ci]|
### Chores
- chore: update npm package to v0.39.5|
- chore: release v0.39.4|
- chore: update npm package to v0.39.4|
- chore: release v0.39.3|
- chore: update npm package to v0.39.3|
- chore: update version to 0.1.1 and add release date to changelog|
- chore(deps): bump the cargo-monthly-rollup group across 1 directory with 28 updates|
- chore: release v0.39.2|
- chore: update npm package to v0.39.2|
# [Version 0.39.4] - 2025-11-03$'
'### Features$'
' - feat: Enhance development and release process for VT Code extension$'
'### Bug Fixes$'
' - fix: rename VT Code Update Plan tool for consistency
- fix: update language model tool properties for VT Code Update Plan$'
'### Refactors$'
' - refactor: remove unused IdeContextBridge and clean up session initialization$'
'### Documentation$'
' - docs: update changelog for v0.39.3 [skip ci]
- docs: update homebrew installation to use core tap
- docs: update changelog for v0.39.2 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.39.4
- chore: release v0.39.3
- chore: update npm package to v0.39.3
- chore: update version to 0.1.1 and add release date to changelog
- chore(deps: bump the cargo-monthly-rollup group across 1 directory with 28 updates
- chore: release v0.39.2
- chore: update npm package to v0.39.2$'
'
# [Version 0.39.3] - 2025-11-03$'
'### Features$'
' - feat: Enhance development and release process for VT Code extension$'
'### Bug Fixes$'
' - fix: rename VT Code Update Plan tool for consistency
- fix: update language model tool properties for VT Code Update Plan$'
'### Documentation$'
' - docs: update changelog for v0.39.2 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.39.3
- chore: update version to 0.1.1 and add release date to changelog
- chore: release v0.39.2
- chore: update npm package to v0.39.2$'
'
# [Version 0.39.2] - 2025-11-03$'
'### Documentation$'
' - docs: update changelog for v0.39.1 [skip ci]
- docs: update tool-policy and extension files to remove quotes from schema_version
- docs: update zed-acp documentation to clarify top-level metadata requirements in extension manifest
- docs: update zed-acp documentation to emphasize required schema_version in extension manifest$'
'### Chores$'
' - chore: update npm package to v0.39.2
- chore: release v0.39.1
- chore: update npm package to v0.39.1$'
'
# [Version 0.39.1] - 2025-11-03$'
'### Documentation$'
' - docs: update tool-policy and extension files to remove quotes from schema_version
- docs: update zed-acp documentation to clarify top-level metadata requirements in extension manifest
- docs: update zed-acp documentation to emphasize required schema_version in extension manifest
- docs: update README and zed-acp documentation to include package.id requirement
- docs: update changelog for v0.39.0 [skip ci]
- docs: update changelog for v0.38.2 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.39.1
- chore: release v0.39.0
- chore: update npm package to v0.39.0
- chore: release v0.38.2
- chore: update npm package to v0.38.2
- chore: update mcp-types integration and add tests for docs.rs compatibility
- chore: update dependencies for agent-client-protocol and related packages
- chore: add sudo to softwareupdate command for OpenSSL installation on macOS
- chore: enhance OpenSSL installation step for x86_64-apple-darwin target$'
'
# [Version 0.39.0] - 2025-11-03$'
'### Features$'
' - feat: Add clear screen command to session and implement related functionality$'
'### Documentation$'
' - docs: update changelog for v0.38.2 [skip ci]
- docs: update changelog for v0.38.1 [skip ci]
- docs: update changelog for v0.38.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.39.0
- chore: release v0.38.2
- chore: update npm package to v0.38.2
- chore: update mcp-types integration and add tests for docs.rs compatibility
- chore: update dependencies for agent-client-protocol and related packages
- chore: add sudo to softwareupdate command for OpenSSL installation on macOS
- chore: enhance OpenSSL installation step for x86_64-apple-darwin target
- chore: release v0.38.1
- chore: update npm package to v0.38.1
- chore: update CI workflow to use stable Rust toolchain and add markdown linting filter
- chore: update dependabot configuration to monthly schedule and reduce open pull requests limit
- chore: release v0.38.0
- chore: update npm package to v0.38.0$'
'
# [Version 0.38.2] - 2025-11-02$'
'### Features$'
' - feat: Add clear screen command to session and implement related functionality$'
'### Documentation$'
' - docs: update changelog for v0.38.1 [skip ci]
- docs: update changelog for v0.38.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.38.2
- chore: update mcp-types integration and add tests for docs.rs compatibility
- chore: update dependencies for agent-client-protocol and related packages
- chore: add sudo to softwareupdate command for OpenSSL installation on macOS
- chore: enhance OpenSSL installation step for x86_64-apple-darwin target
- chore: release v0.38.1
- chore: update npm package to v0.38.1
- chore: update CI workflow to use stable Rust toolchain and add markdown linting filter
- chore: update dependabot configuration to monthly schedule and reduce open pull requests limit
- chore: release v0.38.0
- chore: update npm package to v0.38.0$'
'
# [Version 0.38.1] - 2025-11-02$'
'### Features$'
' - feat: Add clear screen command to session and implement related functionality
- feat: Enhance glob pattern matching to support question mark wildcard
- feat: Enhance tool policy and add time conversion functions$'
'### Bug Fixes$'
' - fix: update tool name in test and improve conversation compression logic$'
'### Refactors$'
' - refactor(file_ops: Optimize file metadata retrieval and reduce unnecessary system calls
- refactor(sandbox: Improve sandbox configuration and event logging performance
- refactor(ui: Modernize TUI rendering and improve diff visualization
- refactor: Improve code formatting and readability in various files
- refactor: update tool policies for curl and apply_patch, and improve error messages in update checker
- refactor: update tool policies to allow more actions and improve asset URL resolution
- refactor: enhance conversation compression logic and message truncation
- refactor: clean up whitespace and improve code readability$'
'### Documentation$'
' - docs: update changelog for v0.38.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.38.1
- chore: update CI workflow to use stable Rust toolchain and add markdown linting filter
- chore: update dependabot configuration to monthly schedule and reduce open pull requests limit
- chore: release v0.38.0
- chore: update npm package to v0.38.0$'
'
# [Version 0.38.0] - 2025-11-02$'
'### Features$'
' - feat: Add clear screen command to session and implement related functionality
- feat: Enhance glob pattern matching to support question mark wildcard
- feat: Enhance tool policy and add time conversion functions$'
'### Bug Fixes$'
' - fix: update tool name in test and improve conversation compression logic
- fix: correct tool name from run_command to run_pty_cmd$'
'### Refactors$'
' - refactor(file_ops: Optimize file metadata retrieval and reduce unnecessary system calls
- refactor(sandbox: Improve sandbox configuration and event logging performance
- refactor(ui: Modernize TUI rendering and improve diff visualization
- refactor: Improve code formatting and readability in various files
- refactor: update tool policies for curl and apply_patch, and improve error messages in update checker
- refactor: update tool policies to allow more actions and improve asset URL resolution
- refactor: enhance conversation compression logic and message truncation
- refactor: clean up whitespace and improve code readability
- refactor: update LLM provider and model configurations
- refactor: rename RUN_PTY_CMD to maintain consistency with run_pty_cmd tool$'
'### Documentation$'
' - docs: update changelog for v0.37.1 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.38.0
- chore: release v0.37.1
- chore: update npm package to v0.37.1
- chore: update dependencies and enhance tool execution reporting$'
'
# [Version 0.37.1] - 2025-10-30$'
'### Features$'
' - feat: enhance command execution policies and UI interactions
- feat: Implement task plan management in TUI session$'
'### Bug Fixes$'
' - fix: correct tool name from run_command to run_pty_cmd
- fix: add Debug trait to InlineTextStyle for improved logging$'
'### Refactors$'
' - refactor: update LLM provider and model configurations
- refactor: rename RUN_PTY_CMD to maintain consistency with run_pty_cmd tool
- refactor: improve tool summary rendering and clean up unused code
- refactor: update LLM provider and model configurations
- refactor: update configuration for LLM provider and model settings
- refactor: move display_interrupt_notice function to improve code organization$'
'### Documentation$'
' - docs: update changelog for v0.37.0 [skip ci]
- docs: update changelog for v0.36.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.37.1
- chore: update dependencies and enhance tool execution reporting
- chore: release v0.37.0
- chore: update npm package to v0.37.0
- chore: release v0.36.0
- chore: update npm package to v0.36.0$'
'
# [Version 0.37.0] - 2025-10-30$'
'### Features$'
' - feat: enhance command execution policies and UI interactions
- feat: Implement task plan management in TUI session
- feat: add asset synchronization script for managing embedded assets
- feat: add embedded asset management for prompts and documentation
- feat: increase max_tool_loops to 100 and add workspace config refresh functionality
- feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: add Debug trait to InlineTextStyle for improved logging
- fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Refactors$'
' - refactor: improve tool summary rendering and clean up unused code
- refactor: update LLM provider and model configurations
- refactor: update configuration for LLM provider and model settings
- refactor: move display_interrupt_notice function to improve code organization$'
'### Documentation$'
' - docs: update changelog for v0.36.0 [skip ci]
- docs: update changelog for v0.35.19 [skip ci]
- docs: add asset synchronization guide for managing embedded assets in vtcode-core
- docs: clean up modules/vtcode_docs_map.md and remove unnecessary newlines in generate-agent-file.md
- docs: update changelog for v0.35.18 [skip ci]
- docs: update changelog for v0.35.17 [skip ci]
- docs: update changelog for v0.35.16 [skip ci]
- docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.37.0
- chore: release v0.36.0
- chore: update npm package to v0.36.0
- chore: release v0.35.19
- chore: update npm package to v0.35.19
- chore: update vtcode and related packages to v0.35.18
- chore: release v0.35.18
- chore: update npm package to v0.35.18
- chore: release v0.35.17
- chore: update npm package to v0.35.17
- chore: release v0.35.16
- chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps: bump crossterm from 0.27.0 to 0.28.1$'
'
# [Version 0.36.0] - 2025-10-30$'
'### Features$'
' - feat: enhance command execution policies and UI interactions
- feat: Implement task plan management in TUI session
- feat: add asset synchronization script for managing embedded assets
- feat: add embedded asset management for prompts and documentation
- feat: increase max_tool_loops to 100 and add workspace config refresh functionality
- feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: add Debug trait to InlineTextStyle for improved logging
- fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Refactors$'
' - refactor: improve tool summary rendering and clean up unused code
- refactor: update LLM provider and model configurations
- refactor: update configuration for LLM provider and model settings
- refactor: move display_interrupt_notice function to improve code organization$'
'### Documentation$'
' - docs: update changelog for v0.35.19 [skip ci]
- docs: add asset synchronization guide for managing embedded assets in vtcode-core
- docs: clean up modules/vtcode_docs_map.md and remove unnecessary newlines in generate-agent-file.md
- docs: update changelog for v0.35.18 [skip ci]
- docs: update changelog for v0.35.17 [skip ci]
- docs: update changelog for v0.35.16 [skip ci]
- docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.36.0
- chore: release v0.35.19
- chore: update npm package to v0.35.19
- chore: update vtcode and related packages to v0.35.18
- chore: release v0.35.18
- chore: update npm package to v0.35.18
- chore: release v0.35.17
- chore: update npm package to v0.35.17
- chore: release v0.35.16
- chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps: bump crossterm from 0.27.0 to 0.28.1$'
'
# [Version 0.35.19] - 2025-10-27$'
'### Features$'
' - feat: add asset synchronization script for managing embedded assets
- feat: add embedded asset management for prompts and documentation
- feat: increase max_tool_loops to 100 and add workspace config refresh functionality
- feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Documentation$'
' - docs: add asset synchronization guide for managing embedded assets in vtcode-core
- docs: clean up modules/vtcode_docs_map.md and remove unnecessary newlines in generate-agent-file.md
- docs: update changelog for v0.35.18 [skip ci]
- docs: update changelog for v0.35.17 [skip ci]
- docs: update changelog for v0.35.16 [skip ci]
- docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.19
- chore: update vtcode and related packages to v0.35.18
- chore: release v0.35.18
- chore: update npm package to v0.35.18
- chore: release v0.35.17
- chore: update npm package to v0.35.17
- chore: release v0.35.16
- chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps: bump crossterm from 0.27.0 to 0.28.1
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.18] - 2025-10-27$'
'### Features$'
' - feat: add embedded asset management for prompts and documentation
- feat: increase max_tool_loops to 100 and add workspace config refresh functionality
- feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Documentation$'
' - docs: update changelog for v0.35.17 [skip ci]
- docs: update changelog for v0.35.16 [skip ci]
- docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.18
- chore: release v0.35.17
- chore: update npm package to v0.35.17
- chore: release v0.35.16
- chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.17] - 2025-10-27$'
'### Features$'
' - feat: increase max_tool_loops to 100 and add workspace config refresh functionality
- feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Documentation$'
' - docs: update changelog for v0.35.16 [skip ci]
- docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.17
- chore: release v0.35.16
- chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.16] - 2025-10-27$'
'### Features$'
' - feat: add templates for agent file generation and VT Code session initiation$'
'### Bug Fixes$'
' - fix: adjust max_tool_loops to 20 and correct prompt file paths$'
'### Documentation$'
' - docs: update changelog for v0.35.15 [skip ci]
- docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.16
- chore: release v0.35.15
- chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.15] - 2025-10-27$'
'### Features$'
' - feat: add templates for agent file generation and VT Code session initiation$'
'### Documentation$'
' - docs: update changelog for v0.35.14 [skip ci]
- docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.15
- chore: release v0.35.14
- chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.14] - 2025-10-27$'
'### Features$'
' - feat: add templates for agent file generation and VT Code session initiation$'
'### Documentation$'
' - docs: update changelog for v0.35.13 [skip ci]
- docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.14
- chore: release v0.35.13
- chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.13] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.12 [skip ci]
- docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.13
- chore: release v0.35.12
- chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.12] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.11 [skip ci]
- docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.12
- chore: release v0.35.11
- chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.11] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.10 [skip ci]
- docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.11
- chore: release v0.35.10
- chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.10] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.9 [skip ci]
- docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.10
- chore: release v0.35.9
- chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.9] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.8 [skip ci]
- docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.9
- chore: release v0.35.8
- chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.8] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.7 [skip ci]
- docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.8
- chore: release v0.35.7
- chore: update npm package to v0.35.7
- chore(deps-dev: bump @typescript-eslint/eslint-plugin
- chore(deps: bump reqwest from 0.12.23 to 0.12.24
- chore(deps: bump is-terminal from 0.4.16 to 0.4.17
- chore(deps: bump parking_lot from 0.12.4 to 0.12.5
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore(deps-dev: bump mocha from 10.8.2 to 11.7.4 in /vscode-extension
- chore(deps-dev: bump @types/glob in /vscode-extension
- chore(deps: bump thiserror from 1.0.69 to 2.0.16
- chore(deps: bump actions/checkout from 4 to 5
- chore(deps: bump DavidAnson/markdownlint-cli2-action from 17 to 20
- chore(deps: bump actions/upload-pages-artifact from 3 to 4$'
'
# [Version 0.35.7] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.6 [skip ci]
- docs: update changelog for v0.35.5 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.7
- chore: release v0.35.6
- chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5$'
'
# [Version 0.35.6] - 2025-10-27$'
'### Features$'
' - feat(minimax: Add MiniMax provider integration and related constants
- feat: update custom prompt command syntax from /prompts to /prompt
- feat: Update README and documentation for Cursor and Windsurf support
- feat: Implement file tree structure for file navigation$'
'### Bug Fixes$'
' - fix(configuration: Update LLM provider and related settings to use OpenRouter
- fix(minimax: Correct base URL in MinimaxProvider configuration
- fix: remove unnecessary newline in CI workflow
- fix: add permissions section to workflow files
- fix: add missing API key header in generate_stream method$'
'### Documentation$'
' - docs: update changelog for v0.35.5 [skip ci]
- docs: update changelog for v0.35.4 [skip ci]
- docs: update changelog for v0.35.3 [skip ci]
- docs: update changelog for v0.35.2 [skip ci]
- docs: update user guide and changelog with quick access shortcuts and enhancements
- docs: update changelog for v0.35.1 [skip ci]
- docs: update changelog for v0.35.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.6
- chore: release v0.35.5
- chore: update npm package to v0.35.5
- chore: release v0.35.4
- chore: update npm package to v0.35.4
- chore: remove example files for self-update and update informer demo
- chore: release v0.35.3
- chore: update npm package to v0.35.3
- chore: remove VSCode extension publishing step from release script
- chore: release v0.35.2
- chore: update npm package to v0.35.2
- chore: release v0.35.1
- chore: update npm package to v0.35.1
- chore: release v0.35.0
- chore: update npm package to v0.35.0
- chore: remove .vscodeignore file and update VSIX package$'
'
# [Version 0.35.5] - 2025-10-27$'
'### Features$'
' - feat(minimax: Add MiniMax provider integration and related constants
- feat: update custom prompt command syntax from /prompts to /prompt
- feat: Update README and documentation for Cursor and Windsurf support
- feat: Implement file tree structure for file navigation$'
'### Bug Fixes$'
' - fix(configuration: Update LLM provider and related settings to use OpenRouter
- fix(minimax: Correct base URL in MinimaxProvider configuration
- fix: remove unnecessary newline in CI workflow
- fix: add permissions section to workflow files
- fix: add missing API key header in generate_stream method$'
'### Documentation$'
' - docs: update changelog for v0.35.4 [skip ci]
- docs: update changelog for v0.35.3 [skip ci]
- docs: update changelog for v0.35.2 [skip ci]
- docs: update user guide and changelog with quick access shortcuts and enhancements
- docs: update changelog for v0.35.1 [skip ci]
- docs: update changelog for v0.35.0 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.5
- chore: release v0.35.4
- chore: update npm package to v0.35.4
- chore: remove example files for self-update and update informer demo
- chore: release v0.35.3
- chore: update npm package to v0.35.3
- chore: remove VSCode extension publishing step from release script
- chore: release v0.35.2
- chore: update npm package to v0.35.2
- chore: release v0.35.1
- chore: update npm package to v0.35.1
- chore: release v0.35.0
- chore: update npm package to v0.35.0
- chore: remove .vscodeignore file and update VSIX package$'
'
# [Version 0.35.4] - 2025-10-27$'
'### Features$'
' - feat(minimax: Add MiniMax provider integration and related constants
- feat: update custom prompt command syntax from /prompts to /prompt
- feat: Update README and documentation for Cursor and Windsurf support
- feat: Implement file tree structure for file navigation$'
'### Bug Fixes$'
' - fix(configuration: Update LLM provider and related settings to use OpenRouter
- fix(minimax: Correct base URL in MinimaxProvider configuration
- fix: remove unnecessary newline in CI workflow
- fix: add permissions section to workflow files
- fix: add missing API key header in generate_stream method$'
'### Documentation$'
' - docs: update changelog for v0.35.3 [skip ci]
- docs: update changelog for v0.35.2 [skip ci]
- docs: update user guide and changelog with quick access shortcuts and enhancements
- docs: update changelog for v0.35.1 [skip ci]
- docs: update changelog for v0.35.0 [skip ci]
- docs: update changelog for v0.35.3 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.4
- chore: remove example files for self-update and update informer demo
- chore: release v0.35.3
- chore: update npm package to v0.35.3
- chore: remove VSCode extension publishing step from release script
- chore: release v0.35.2
- chore: update npm package to v0.35.2
- chore: release v0.35.1
- chore: update npm package to v0.35.1
- chore: release v0.35.0
- chore: update npm package to v0.35.0
- chore: remove .vscodeignore file and update VSIX package
- chore: release v0.35.3
- chore: update npm package to v0.35.3
- chore: remove VSCode extension publishing step from release script$'
'
# [Version 0.35.3] - 2025-10-27$'
'### Documentation$'
' - docs: update changelog for v0.35.2 [skip ci]
- docs: update user guide and changelog with quick access shortcuts and enhancements$'
'### Chores$'
' - chore: update npm package to v0.35.3
- chore: remove VSCode extension publishing step from release script
- chore: release v0.35.2
- chore: update npm package to v0.35.2$'
'
# [Version 0.35.2] - 2025-10-27$'
'### Documentation$'
' - docs: update user guide and changelog with quick access shortcuts and enhancements
- docs: update changelog for v0.35.1 [skip ci]$'
'### Chores$'
' - chore: update npm package to v0.35.2
- chore: release v0.35.1
- chore: update npm package to v0.35.1$'
'
## [0.35.1] - 2025-10-27
### Features
- feat: update custom prompt command syntax from /prompts to /prompt
- feat: Update README and documentation for Cursor and Windsurf support
- feat: Implement file tree structure for file navigation
- feat: add simple GitHub Pages workflow for /docs
- feat: Enhance model picker and dynamic model fetching
- feat: add synchronous fetching of LMStudio models and improve model selection
### Bug Fixes
- fix: remove mdbook workflow causing CI failure
### Documentation
- docs: update changelog for v0.35.0 [skip ci]
- docs: add lifecycle hooks guide
### Chores
- chore: update npm package to v0.35.1
- chore: release v0.35.0
- chore: update npm package to v0.35.0
- chore: remove .vscodeignore file and update VSIX package
- chore(deps-dev): bump esbuild in /vscode-extension
- chore(deps-dev): bump eslint from 8.57.1 to 9.38.0 in /vscode-extension
- chore(deps): bump windows-sys from 0.59.0 to 0.61.1
- chore(deps): bump toml from 0.9.7 to 0.9.8
- chore(deps): bump tree-sitter-javascript from 0.23.1 to 0.25.0
- chore(deps): bump dirs from 5.0.1 to 6.0.0
- chore(deps-dev): bump @types/node in /vscode-extension
- chore(deps): bump tree-sitter-go from 0.23.4 to 0.25.0
- chore(deps-dev): bump glob from 10.4.5 to 11.0.3 in /vscode-extension
- chore(deps): bump actions/cache from 3 to 4
- chore(deps): bump actions/upload-artifact from 4 to 5
- chore(deps-dev): bump @typescript-eslint/parser in /vscode-extension
- chore(deps): bump codecov/codecov-action from 3 to 5
- chore(deps): bump actions/checkout from 3 to 5
- chore(deps): bump actions/setup-node from 4 to 6
## [0.35.0] - 2025-10-27
### Features
- feat: update custom prompt command syntax from /prompts to /prompt
- feat: Update README and documentation for Cursor and Windsurf support
- feat: Implement file tree structure for file navigation
- feat: add simple GitHub Pages workflow for /docs
- feat: Enhance model picker and dynamic model fetching
- feat: add synchronous fetching of LMStudio models and improve model selection
- feat: Add IDE integration and troubleshooting guides to documentation
- feat: Add VSCode extension publishing support to release script
- feat: Add initial files for VT Code Companion extension including README, LICENSE, CHANGELOG, and esbuild configuration
- feat: Add initial package.json for VT Code Companion extension
- feat(security): Implement comprehensive security documentation and fixes
- feat: add comprehensive security audit and model documentation
### Bug Fixes
- fix: remove mdbook workflow causing CI failure
### Refactors
- refactor: Rename extension from "VT Code Companion" to "VT Code" and update CHANGELOG
- refactor: use unsafe blocks for environment variable manipulation in tests
- refactor: remove unused tools and simplify tool policies
### Documentation
- docs: update changelog for v0.34.0 [skip ci]
- docs: add lifecycle hooks guide
### Chores
- chore: update npm package to v0.35.0
- chore: remove .vscodeignore file and update VSIX package
- chore(deps-dev): bump esbuild in /vscode-extension
- chore(deps-dev): bump eslint from 8.57.1 to 9.38.0 in /vscode-extension
- chore(deps): bump windows-sys from 0.59.0 to 0.61.1
- chore(deps): bump toml from 0.9.7 to 0.9.8
- chore(deps): bump tree-sitter-javascript from 0.23.1 to 0.25.0
- chore(deps): bump dirs from 5.0.1 to 6.0.0
- chore(deps-dev): bump @types/node in /vscode-extension
- chore(deps): bump tree-sitter-go from 0.23.4 to 0.25.0
- chore(deps-dev): bump glob from 10.4.5 to 11.0.3 in /vscode-extension
- chore(deps): bump actions/cache from 3 to 4
- chore(deps): bump actions/upload-artifact from 4 to 5
- chore(deps-dev): bump @typescript-eslint/parser in /vscode-extension
- chore(deps): bump codecov/codecov-action from 3 to 5
- chore(deps): bump actions/checkout from 3 to 5
- chore(deps): bump actions/setup-node from 4 to 6
- chore: release v0.34.0
- chore: update npm package to v0.34.0
## [0.34.0] - 2025-10-25
### Features
- feat: Add IDE integration and troubleshooting guides to documentation
- feat: Add VSCode extension publishing support to release script
- feat: Add initial files for VT Code Companion extension including README, LICENSE, CHANGELOG, and esbuild configuration
- feat: Add initial package.json for VT Code Companion extension
- feat(security): Implement comprehensive security documentation and fixes
- feat: add comprehensive security audit and model documentation
- feat: add changelog generation from commits in release script
### Refactors
- refactor: Rename extension from "VT Code Companion" to "VT Code" and update CHANGELOG
- refactor: use unsafe blocks for environment variable manipulation in tests
- refactor: remove unused tools and simplify tool policies
### Documentation
- docs: update changelog for v0.33.1 [skip ci]
### Chores
- chore: update npm package to v0.34.0
- chore: release v0.33.1
- chore: update npm package to v0.33.1
- chore: update README.md for improved installation instructions and feature highlights
- chore: update CHANGELOG.md with recent enhancements for v0.33.0
## [0.33.1] - 2025-01-30
### Features
- feat: add changelog generation from commits in release script
- feat: run doctests separately in publish_extracted_crates.sh
- feat: add comprehensive plan for open sourcing VT Code core components
- feat: add demo section with updated demo GIF in README
- feat: add VT Code VHS showcase and demo files
### Chores
- chore: update npm package to v0.33.1
- chore: update README.md for improved installation instructions and feature highlights
- chore: update CHANGELOG.md with recent enhancements for v0.33.0
- chore: release v0.33.0
- chore: update npm package to v0.33.0
- chore: update package versions to 0.32.0 and adjust dependencies
- chore: update npm package to v0.32.0
- chore: update demo GIF for VHS showcase
### Recent Enhancements (v0.33.0 and beyond)
- **Enhanced Tool Execution & Output Handling**: Improved tool execution with better error handling and output formatting for enhanced reliability and user experience
- **Enhanced Timeout Detection & Token Budget Management**: Improved timeout handling and more sophisticated token budget management with better attention management for enhanced performance
- **Improved Output Rendering**: Enhanced syntax highlighting for JSON, XML, and YAML outputs with better error messaging
- **Enhanced Bash Runner & Telemetry**: Added dry-run capabilities and feature-gated executors for shell operations with integrated telemetry
- **Ollama Integration Improvements**: Better support for local models with configurable base URLs and improved tool call handling
- **MCP Protocol & Tool Support**: Enhanced Model Context Protocol integration with improved resource and prompt handling
- **Configuration System Improvements**: Enhanced configuration handling with better default preservation and schema validation
- **Component Extraction Strategy**: Continued work on extracting reusable components including vtcode-exec-events, vtcode-bash-runner, vtcode-config, and vtcode-indexer
### Extracted crates release preparation
- **vtcode-commons 0.1.0** – marks the shared workspace path/telemetry traits crate ready for publishing with repository and
documentation metadata in `Cargo.toml`.
- **vtcode-markdown-store 0.1.0** – aligns the markdown-backed storage crate with the initial release version and links to the
public documentation.
- **vtcode-indexer 0.1.0** – retags the workspace-friendly indexer for its first standalone release and records the docs.rs URL
for consumers.
- **vtcode-bash-runner 0.1.0** – updates the shell execution helper crate to the shared release version, adds licensing
metadata, and points to hosted documentation.
- **vtcode-exec-events 0.1.0** – finalizes the telemetry schema crate for release with docs.rs metadata alongside the version
alignment.
- Ran `cargo publish --dry-run` for the release candidates (`vtcode-commons`, `vtcode-markdown-store`, `vtcode-indexer`, `vtcode-exec-events`) and confirmed that `vtcode-bash-runner` will package successfully once `vtcode-commons` is available on crates.io.
- Scheduled the sequential publish order, tagging plan, and post-release dependency bumps in `docs/component_release_plan.md` so the crates can be released without coordination gaps.
- Scripted the sequential publish workflow in `scripts/publish_extracted_crates.sh` to automate validation, publishing, and tagging steps with optional dry-run rehearsals.
### `vtcode-exec-events`
- Added schema metadata (`EVENT_SCHEMA_VERSION`) and a `VersionedThreadEvent` wrapper so consumers can negotiate compatibility before processing telemetry streams.
- Introduced an `EventEmitter` trait with optional `LogEmitter` and `TracingEmitter` adapters to integrate JSON and tracing pipelines without boilerplate.
- Published JSON helper utilities and optional schema export support to simplify serialization round-trips and documentation workflows.
### `vtcode-bash-runner`
- Added feature-gated executors for process, pure-Rust, and dry-run operation so adopters can tailor shell execution strategies without forking the runner.F:vtcode-bash-runner/Cargo.toml†L1-L40F:vtcode-bash-runner/src/executor.rs†L1-L356
- Introduced the `EventfulExecutor` bridge to emit `vtcode-exec-events` telemetry from standalone shell invocations, plus documentation covering the new feature flags and integrations.F:vtcode-bash-runner/src/executor.rs†L358-L470F:docs/modules/modules/vtcode_bash_runner.md†L1-L120F:docs/modules/modules/vtcode_exec_events.md†L1-L160
### **Major Enhancements - Context Engineering & Attention Management** (Phase 1 & 2)
#### Phase 1: Enhanced System Prompts
- **Explicit Response Framework**: All system prompts now include a clear 5-step framework
1. Assess the situation - Understand what the user needs
2. Gather context efficiently - Use search tools before reading files
3. Make precise changes - Prefer targeted edits over rewrites
4. Verify outcomes - Test changes appropriately
5. Confirm completion - Summarize and verify satisfaction
- **Enhanced Guidelines**: More specific guidance on tool selection, code style preservation, and handling destructive operations
- **Multi-Turn Coherence**: Explicit guidance on building context across conversation turns
- **Token Efficiency**: Maintained concise prompts (~280 tokens) while adding structure
**System Prompt Improvements:**
- Default prompt: Added explicit framework, guidelines, and context management strategies
- Lightweight prompt: Added minimal 4-step approach for quick tasks
- Specialized prompt: Added tool selection strategy by phase, advanced guidelines, and multi-turn coherence
#### Phase 2: Dynamic Context Curation
- **New Module**: `context_curator.rs` - Implements iterative per-turn context selection based on Anthropic's principles
- **Conversation Phase Detection**: Automatically detects phase (Exploration, Implementation, Validation, Debugging, Unknown)
- **Phase-Aware Tool Selection**: Dynamically selects relevant tools based on current conversation phase
- **Priority-Based Context Selection**:
1. Recent messages (always included, configurable)
2. Active work context (files being modified)
3. Decision ledger summary (compact)
4. Recent errors and resolutions
5. Relevant tools (phase-aware)
- **Configurable Curation**: Full control via `[context.curation]` configuration
**Key Features:**
- Tracks active files and file summaries
- Maintains recent error context for debugging
- Selects optimal tools based on conversation phase
- Respects token budget constraints
- Integrates with TokenBudgetManager and DecisionTracker
**API:**
```rust
let curator = ContextCurator::new(config, token_budget, decision_ledger);
curator.mark_file_active("src/main.rs".to_string());
curator.add_error(ErrorContext { ... });
let curated = curator.curate_context(&messages, &tools).await?;
```
**Configuration:**
```toml
[context.curation]
enabled = true
max_tokens_per_turn = 100000
preserve_recent_messages = 5
max_tool_descriptions = 10
include_ledger = true
ledger_max_entries = 12
include_recent_errors = true
max_recent_errors = 3
```
#### Token Budget Tracking & Attention Management
- **New Module**: `token_budget.rs` - Real-time token budget tracking using Hugging Face `tokenizers`
- **Component-Level Tracking**: Monitor token usage by category (system prompt, messages, tool results, decision ledger)
- **Configurable Thresholds**: Warning at 75% (customizable via `vtcode.toml`)
- **Model-Specific Tokenizers**: Support for GPT, Claude, and other models for accurate counting
- **Automatic Deduction**: Track token removal during context cleanup
- **Budget Reports**: Generate detailed token usage reports by component
- **Performance Optimized**: ~10μs per message using Rust-native Hugging Face `tokenizers`
- **New Method**: `remaining_tokens()` - Get remaining tokens in budget for context curation decisions
**Configuration:**
```toml
[context.token_budget]
enabled = true
model = "gpt-5-nano"
warning_threshold = 0.75
detailed_tracking = false
```
#### Optimized System Prompts & Tool Descriptions
- **67-82% Token Reduction**: System prompts streamlined from ~600 tokens to ~200 tokens
- **80% Tool Description Efficiency**: Average tool description reduced from ~400 to ~80 tokens
- **"Right Altitude" Principles**: Concise, actionable guidance over verbose instructions
- **Progressive Disclosure**: Emphasize search-first approach with `grep_file`
- **Clear Tool Purposes**: Eliminated capability overlap in tool descriptions
- **Token Management Guidance**: Built-in advice for efficient context usage (e.g., `max_results` parameters)
**System Prompt Improvements:**
- Removed verbose explanations and redundant information
- Focused on core principles and actionable strategies
- Added explicit context strategy guidance
- Emphasized metadata-first, content-second approach
**Tool Description Improvements:**
- Clear, unambiguous purposes with minimal overlap
- Token efficiency guidance (e.g., `max_results` limits)
- Auto-chunking behavior documented
- Metadata-first approach emphasized
#### Context Engineering Documentation
- **New Documentation**: `docs/context_engineering.md` - Comprehensive guide to context management
- **Implementation Summary**: `docs/context_engineering_implementation.md` - Technical details
- **Best Practices**: User and developer guidelines for efficient context usage
- **Configuration Examples**: Complete examples for token budget and context management
- **Performance Metrics**: Token efficiency improvements documented
- **References**: Links to Anthropic research and related resources
#### Bug Fixes
- **Fixed MCP Server Initialization**: Removed premature `cleanup_dead_providers()` call that caused `BrokenPipeError` during initialization
- **MCP Process Management**: Improved connection lifecycle management to prevent pipe closure issues
#### Dependencies
- **Added**: `tokenizers = "0.15"` for accurate token counting
- **Updated**: Cargo.lock with new dependencies
#### Release Automation
- **Cargo Release Integration**: Adopted `cargo release` with a shared workspace configuration (`release.toml`) and updated `scripts/release.sh` to drive changelog-powered GitHub releases, coordinated crates.io publishing, and npm version synchronization.
### **Major Enhancements - Anthropic-Inspired Architecture**
#### Decision Transparency System
- **New Module**: `decision_tracker.rs` - Complete audit trail of all agent decisions
- **Real-time Tracking**: Every action logged with reasoning and confidence scores
- **Transparency Reports**: Live decision summaries and session statistics
- **Confidence Scoring**: Quality assessment for all agent actions
- **Context Preservation**: Full conversation context maintained across decisions
#### Error Recovery & Resilience
- **New Module**: `error_recovery.rs` - Intelligent error handling system
- **Pattern Detection**: Automatic identification of recurring errors
- **Context Preservation**: Never lose important information during failures
- **Recovery Strategies**: Multiple approaches for handling errors gracefully
- **Error Statistics**: Comprehensive analysis of error patterns and recovery rates
#### Conversation Summarization
- **New Module**: `conversation_summarizer.rs` - Automatic conversation compression
- **Intelligent Summaries**: Key decisions, completed tasks, and error patterns
- **Long Session Support**: Automatic triggers when conversations exceed thresholds
- **Confidence Scoring**: Quality assessment for summary reliability
- **Context Efficiency**: Maintain useful context without hitting limits
### **Tool Design Improvements**
#### Enhanced Tool Documentation
- **Comprehensive Specifications**: Extensive tool descriptions with examples and error cases
- **Error-Proofing**: Anticipate and prevent common model misunderstandings
- **Clear Usage Guidelines**: Detailed instructions for each tool parameter
- **Debugging Support**: Specific guidance for troubleshooting tool failures
#### Improved System Instruction
- **Model-Driven Control**: Give maximum autonomy to the language model
- **Thorough Reasoning**: Encourage deep thinking for complex problems
- **Flexible Methodology**: Adaptable problem-solving approaches
- **Quality First**: Emphasize correctness over speed
### **Release Automation**
- **Coordinated Version Bumps**: `scripts/release.sh` now prompts maintainers to bump the `vtagent-core` crate alongside the main binary, keeping release metadata synchronized.
### **Transparency & Observability**
#### Verbose Mode Enhancements
- **Real-time Decision Tracking**: See exactly why each action is taken
- **Error Recovery Monitoring**: Observe intelligent error handling
- **Conversation Summarization Alerts**: Automatic notifications for long sessions
- **Session Statistics**: Comprehensive metrics and pattern analysis
- **Pattern Detection**: Automatic identification of recurring issues
#### Session Reporting
- **Final Transparency Reports**: Complete session summaries with success metrics
- **Error Recovery Statistics**: Analysis of error patterns and recovery rates
- **Decision Quality Metrics**: Confidence scores and decision success rates
- **Context Usage Monitoring**: Automatic warnings for approaching limits
### **Configuration System Improvements**
#### Two-Way Configuration Synchronization
- **Smart Config Generation**: `vtcode config` now reads existing `vtcode.toml` and preserves customizations
- **Complete Template Generation**: Ensures all configuration sections are present, even missing ones
- **Bidirectional Sync**: Generated configs always match your actual configuration state
- **Fallback Safety**: Uses system defaults when no configuration file exists
- **TOML Serialization**: Replaced hardcoded templates with proper TOML generation
## [Previous Versions]
### v0.1.0 - Initial Release
- Basic agent architecture with Gemini integration
- Core file system tools (list_files, read_file, write_file, edit_file)
- Interactive chat and specialized workflows
- Workspace safety and path validation
- Comprehensive logging and debugging support
## **Performance & Reliability**
### SWE-bench Inspired Improvements
- **49% Target Achievement**: Architecture designed following Anthropic's breakthrough approach
- **Error-Proofed Tools**: Extensive validation and error handling
- **Context Engineering**: Research-preview conversation management techniques
- **Model Empowerment**: Maximum control given to language models
### Reliability Enhancements
- **Context Preservation**: Never lose important information during failures
- **Recovery Strategies**: Multiple approaches for error handling
- **Pattern Detection**: Automatic identification of recurring issues
- **Comprehensive Logging**: Full audit trail of all agent actions
## **Technical Improvements**
### Architecture Refactoring
- **Modular Design**: Separate modules for transparency, error recovery, and summarization
- **Clean Interfaces**: Well-defined APIs between components
- **Performance Optimization**: Efficient data structures and algorithms
- **Error Handling**: Comprehensive error management throughout
### Code Quality
- **Documentation**: Extensive inline documentation and examples
- **Type Safety**: Strong typing with comprehensive error handling
- **Testing**: Unit tests for core functionality
- **Linting**: Clean, well-formatted code following Rust best practices
## **Key Features Summary**
### New Capabilities
1. **Complete Decision Transparency** - Every action tracked and explained
2. **Intelligent Error Recovery** - Learn from mistakes and adapt strategies
3. **Confidence Scoring** - Quality assessment for all agent actions
4. **Pattern Detection** - Identify and address recurring issues
### Enhanced User Experience
1. **Verbose Mode Overhaul** - Rich transparency and debugging information
2. **Better Error Messages** - Clear, actionable feedback for all failures
3. **Session Insights** - Comprehensive statistics and recommendations
4. **Improved Tool Reliability** - Error-proofed design prevents common issues
5. **Context Management** - Intelligent handling of conversation limits
## **Future Roadmap**
### Planned Enhancements
- **Multi-file Operations**: Batch processing capabilities
- **Project Templates**: Predefined scaffolds for common projects
- **Integration APIs**: REST endpoints for external integration
### Research Areas
- **Multi-modal Support**: Images, diagrams, and audio processing
- **Collaborative Workflows**: Enhanced human-agent teaming
- **Domain Specialization**: Industry-specific optimizations
- **Performance Benchmarking**: SWE-bench style evaluation capabilities
## **Contributing**
### Development Guidelines
- **Feature Branches**: Create feature branches for new capabilities
- **Comprehensive Testing**: Include tests for all new functionality
- **Documentation Updates**: Update README, BUILD.md, and this CHANGELOG
- **Code Standards**: Follow established Rust idioms and best practices
### Areas of Interest
- **Tool Enhancements**: Additional tools for specific use cases
- **Workflow Patterns**: New specialized workflows and patterns
- **Performance Optimization**: Further improvements for complex tasks
- **Documentation**: Tutorials, examples, and user guides
---
## **Related Breakthroughs**
This release incorporates insights from Anthropic's engineering approach that achieved **49% on SWE-bench Verified**, including:
- **Minimal Scaffolding**: Give maximum control to language models
- **Error-Proofed Tools**: Extensive documentation and validation
- **Thorough Reasoning**: Encourage deep thinking for complex problems
- **Context Preservation**: Never lose important information during failures
- **Decision Transparency**: Complete audit trail of agent actions
These improvements position vtcode as a state-of-the-art coding assistant with exceptional transparency, reliability, and performance on complex software engineering tasks.