Crate koda_core 

Koda Core — the engine library for the Koda AI coding agent.

This crate contains the pure engine logic with zero terminal dependencies. It communicates exclusively through EngineEvent (output) and EngineCommand (input) enums.

See DESIGN.md in the repository root for the full architectural rationale.

Modules

agent

Sub-agent configuration, discovery, and invocation. KodaAgent — shared, immutable agent resources.

approval

Tool approval modes, safety gates, and shared mode state. Approval modes and tool confirmation.

bash_path_lint

Heuristic path-escape detection for bash commands. Bash path lint: detect commands that escape the project root.

bash_safety

Bash command safety classification (destructive, mutating, read-only). Bash command safety classification.

compact

Context compaction — summarise old messages to reclaim token budget. Session compaction — summarize old messages to reclaim context.

config

Global configuration: provider, model, model settings, CLI flags. Configuration loading for agents and global settings.

context

Context window management and token budgeting. Context window tracking.

db

SQLite persistence layer — sessions, messages, usage tracking. SQLite persistence layer.

engine

Engine protocol: EngineEvent / EngineCommand enums. Engine module: the protocol boundary between Koda’s core and any client.

git

Git helpers — status, diff, blame, log. Git integration for context injection.

inference

The main inference loop — send messages, stream responses, dispatch tools. LLM inference loop with streaming, tool execution, and sub-agent delegation.

inference_helpers

Shared helpers used by the inference loop. Helper functions for inference — context estimation, message assembly, error classification.

keystore

Credential storage (OS keychain via keyring). Secure API key storage.

loop_guard

Guardrail against runaway tool-call loops. Loop detection and hard-cap user prompt for the inference loop.

mcp

MCP (Model Context Protocol) server connections and tool routing. MCP (Model Context Protocol) support.

memory

Project memory — MEMORY.md / CLAUDE.md read/write. Semantic memory: project context injected into the system prompt.

model_context

Hardcoded context-window lookup table (fallback when API doesn’t report). Model context window lookup.

output_caps

Context-scaled output caps for tool results. Centralized tool output caps, scaled to the model’s context window.

persistence

Persistence trait — the database contract. Persistence trait — the storage contract for koda.

preview

Diff preview generation for file mutations. Pre-confirmation diff previews for destructive tool operations.

progress

Progress reporting helpers for long operations. Structured progress tracking.

prompt

System prompt construction. System prompt construction.

providers

LLM provider abstraction — Anthropic, Gemini, OpenAI-compatible. LLM provider abstraction layer.

runtime_env

Environment variable access (mockable for tests). Thread-safe runtime environment for API keys and config.

session

Session lifecycle — create, resume, list, delete. KodaSession — per-conversation state.

settings

User settings persistence (~/.config/koda/settings.json). User settings persistence.

skills

Skill discovery and activation (project, user, built-in). Skill discovery and loading.

sub_agent_cache

Cache for sub-agent provider/model config across invocations. Sub-agent result caching.

tool_dispatch

Tool dispatch — routes tool calls from inference to the registry. Tool execution dispatch — sequential, parallel, and sub-agent.

tools

Tool registry, definitions, execution, and path safety. Tool registry and execution engine.

truncate

Token-safe output truncation. Output truncation for display.

undo

Undo stack for file mutations. Undo stack for file mutations.

version

Version string and update-check helpers. Version checker: non-blocking startup check for newer crate versions.