llman 0.0.26

A tool for managing LLM application rules(prompts) ...
Documentation
schema: spec-driven

context: |
  Project: llman CLI quality uplift
  Primary usage: interactive CLI distributed via cargo install.
  Platforms: Linux and macOS are primary; Windows support is partial and not a target.
  Compatibility: output and exit codes may change, but changes must be documented per task.

  Constraints:
  - Do not touch real user config in tests/dev commands; use `LLMAN_CONFIG_DIR`.
  - Keep changes incremental and reviewable (one task per PR/merge).

  Goals:
  - Improve maintainability, readability, and separation of concerns.
  - Improve reliability, error signaling, testability, and CI signal quality.
  - Improve CLI experience: error messages, help, and consistency.

  Non-goals:
  - No large new frameworks or rewrites.
  - No full Windows support expansion.
  - No breaking changes without a documented rollback path.

  Guiding principles:
  - Small, mergeable steps (one task at a time).
  - Prefer shared helpers in `src/config.rs` and `src/path_utils.rs`.
  - Fail loudly and consistently for errors.
  - Avoid risky behavior when parsing or modifying user files.

  Risks and mitigations:
  - Output and exit code changes can surprise users; document changes and provide examples.
  - Config path changes can move data; keep the same default path and add migration notes.
  - Safer comment cleaning may remove fewer comments; warn clearly and keep risky fallback opt-in.
  - Stricter CI can slow feedback; keep steps minimal and prefer `just check`.

  Milestones:
  - M1: consistent config path resolution and error/exit handling.
  - M2: cursor export correctness and safer tool behavior.
  - M3: quality gates (fmt/clippy) and message consistency.

  Acceptance overview:
  - `cargo +nightly fmt -- --check` passes.
  - `cargo +nightly clippy --all-targets --all-features -- -D warnings` passes.
  - `cargo +nightly test --all` passes.
  - Manual smoke checks for `llman x cc`, `llman x codex`, `llman x cursor`, `llman prompt`, `llman tool`.

# Project context (optional)
# This is shown to AI when creating artifacts.
# Add your tech stack, conventions, style guides, domain knowledge, etc.
# Example:
#   context: |
#     Tech stack: TypeScript, React, Node.js
#     We use conventional commits
#     Domain: e-commerce platform

# Per-artifact rules (optional)
# Add custom rules for specific artifacts.
# Example:
#   rules:
#     proposal:
#       - Keep proposals under 500 words
#       - Always include a "Non-goals" section
#     tasks:
#       - Break tasks into chunks of max 2 hours