LLMSkillHub
mirroredCopied from dealfluence/adeu under MIT. dealfluence has not claimed this page, and is not affiliated with LLM SkillHub.

dealfluence / hunt-comments

Use when hunting down verbose, narrative, or dead comments in the codebase to reduce comment noise and character counts.

0
SKILL.md
1.4 KB

Hunt Comments Skill

This skill provides a systematic approach to finding and reducing comment density and character bloat across source code files.

Workflow

  1. Run the Comment Density Spotter:
    Execute python scripts/count_comment_chars.py to rank files by comment character count:

     uv run python scripts/count_comment_chars.py -n 25

  2. Target High-Density Files:
    Focus on top-ranked files with high comment ratios (>20% comment characters).

  3. Identify & Remove Non-Essential Commentary:

    • Delete narrative historical explanations ("Why we added this in 2024...", "Fix for issue #123 where...").
    • Strip redundant docstrings that simply restate function signatures, types, or obvious operations.
    • Collapse multi-paragraph block comments into brief 1-line "why" notes or remove them entirely if code is self-explanatory.
    • Retain only essential architectural invariants (e.g., OPC part boundaries, Word signed int32 overflow rules, security guards).

  4. Verify Behavior:
    Always run test suites to verify zero functional or formatting regressions:

    • Python: uv run pytest & uv run mypy src
    • Node: npm run build && npm test
    • LangChain: uv run pytest in langchain