|
| 1 | +--- |
| 2 | +name: docs-cleanup |
| 3 | +description: Audit repository documentation for stale, conflicting, or incomplete install/run guidance and outdated examples; report concrete file-level gaps and recommended fixes. |
| 4 | +--- |
| 5 | + |
| 6 | +# Docs Cleanup |
| 7 | + |
| 8 | +Use this skill when the user asks for a documentation audit, cleanup pass, or stale-doc review. |
| 9 | + |
| 10 | +## Goals |
| 11 | + |
| 12 | +- Find conflicting or outdated install/run instructions. |
| 13 | +- Check whether examples still match current CLI/API behavior. |
| 14 | +- Flag missing guidance that blocks onboarding, testing, or release tasks. |
| 15 | +- Keep output focused on actionable findings with file references. |
| 16 | + |
| 17 | +## Workflow |
| 18 | + |
| 19 | +1. Collect candidate docs: |
| 20 | + - `README.md` |
| 21 | + - `docs/source/**/*.rst` |
| 22 | + - `CONTRIBUTING.md` |
| 23 | + - `.github/` templates/workflows when relevant to user instructions. |
| 24 | +2. Validate install/run instructions against current repo behavior: |
| 25 | + - Build and test commands in docs vs commands in `CMakeLists.txt`, `pyproject.toml`, CI workflows, and `scripts/`. |
| 26 | + - Confirm paths, filenames, and prerequisites are still valid. |
| 27 | +3. Validate examples: |
| 28 | + - Compare documented examples with `samples/`, `test/`, and current interfaces. |
| 29 | + - Check that command arguments and expected outputs are still plausible. |
| 30 | +4. Identify gaps: |
| 31 | + - Missing prerequisites, platform caveats, troubleshooting notes, or release-process notes. |
| 32 | +5. Report findings first, ordered by severity: |
| 33 | + - `High`: wrong instruction likely to fail. |
| 34 | + - `Medium`: confusing/outdated but recoverable. |
| 35 | + - `Low`: clarity/consistency improvements. |
| 36 | + |
| 37 | +## Output Format |
| 38 | + |
| 39 | +- Findings list with `severity`, `file`, and exact issue. |
| 40 | +- Proposed fix for each finding. |
| 41 | +- Open questions or assumptions. |
| 42 | +- If no findings: explicitly state that and note residual risk areas not fully validated. |
| 43 | + |
| 44 | +## Guardrails |
| 45 | + |
| 46 | +- Prefer small documentation edits over broad rewrites. |
| 47 | +- Avoid speculative claims; cite the source file/command that contradicts the docs. |
| 48 | +- Do not edit generated docs under `docs/_build/` or `docs/doxygen/`. |
0 commit comments