doc-audit
Skill3 starsChanged 48 days agoMITView on GitHub
Audit and surgically fix the marketplace's documentation so every shipped command is discoverable — checks each plugin README's Skills section against SKILL.md frontmatter, the root README's plugin list and command index, manifest/description sync, and docs-site parity. Use when docs may have drifted from the code, after adding or renaming a skill or plugin, or before a release.
12 other projects use a version of this.
Adopted by
Most recently taken up by
- blueclover22/mw_agent_kittheir own version, first seen Nov 4, 2026
- florian101010/awesome-agentic-AI-coding-templatetheir own version, first seen Nov 1, 2026
- sunitghub/canon-skillstheir own version, first seen Oct 22, 2026
- Sprintlaw/legal-skillstheir own version, first seen Oct 17, 2026
- jrittelmeyer/ai-dev-kittheir own version, first seen Oct 15, 2026
- bekalpaslan/promptlinetheir own version, first seen Oct 13, 2026
Other versions12 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 12
- PackmindHub/packmind★ 323same name
- tachyon-zcash/ragu★ 61same name
- bekalpaslan/promptline★ 13same name
- sunitghub/canon-skills★ 9same name
- florian101010/awesome-agentic-AI-coding-template★ 4same name
- Wubabalala/claude-skills★ 4same name
- blueclover22/mw_agent_kit★ 1same name
- jarosser06/drift★ 1same name
- jrittelmeyer/ai-dev-kit★ 1same name
- s-frantz/agent-memory-trellis★ 1same name
2 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- feature-doneSkill
- progressive-disclosureSkill
- release-docsSkill
- test-plugins-locallySkill
- validateSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 12 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Never run the fix pass outside a git working tree where changes can be reviewed/reverted.
- Distinctive: line 21
Run the audit and read the full report:
- Very specific: line 36
Read references/completeness-contract.md for the exact rules and scaffold shape.
- Distinctive: line 46
Surgical rule: never rewrite an existing hand-written entry.
- Distinctive: line 72
Leave existing hand-written documentation intact — add, don't rewrite.
“Common” is measured: kept by at least 6 in 10 of the 12 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/doc-audit — Documentation Completeness Auditor & Fixer
Audit the marketplace's docs against machine truth (SKILL.md frontmatter + manifests), then
surgically fix the mechanical gaps and flag the rest. Read-only detection lives in
audit-docs.sh; the edits below are applied by you, the model, only on the fix pass.
Guard
This skill edits files (READMEs). It is user-invoked only. On a real run it will modify
plugin READMEs and the root README in the working tree — the user reviews via git diff.
Never run the fix pass outside a git working tree where changes can be reviewed/reverted.
Step 1: Detect
Run the audit and read the full report:
bash .claude/skills/doc-audit/scripts/audit-docs.sh --format pretty(Optionally --root <dir> to audit a different checkout, e.g. a worktree.)
Step 2: Present the report
Summarise the findings grouped by severity (ERROR / WARN / INFO) and by layer. Include the coverage table (command → documented?) so the user sees included-vs-omitted without asking.
Step 3: Surgical fix pass (auto-fixable findings only)
Read references/completeness-contract.md for the exact rules and scaffold shape. Then, for each
auto-fixable finding, make the smallest edit that resolves it:
MISSING_SKILLS_SECTION→ add a## Skillssection scaffolded from frontmatter.CMD_NOT_IN_README→ insert one entry (scaffold shape) in canonical order.MARKER_MISSING→ add*(destructive — user-invoked only)*to that entry's header only.PLUGIN_NOT_IN_ROOT→ add the plugin to the root "at a glance" table and command index.ROOT_COUNT_STALE→ correct the plugin-count word/number.ROOT_NO_CMD_INDEX→ add a central command index listing every plugin's user commands. If a command section already exists, extend it in place — never add a second index (the fix is idempotent: re-running must produce no further change).
Surgical rule: never rewrite an existing hand-written entry. Only add what is missing or correct a marker/count. Preserve tables, prose, and ordering already present.
Step 4: Flag the rest (no edits)
Report flag-only findings verbatim for the user to handle:
PHANTOM_CMD, VERSION_MISMATCH, DESC_MISMATCH, PLUGIN_JSON_MISSING,
SITE_PAGE_MISSING. For that finding, the instruction is "run /walkthrough:document <plugin>".
Site content drift is checked by /release-docs and its "Docs Obligations" gate, not here.
Step 5: Re-verify
Re-run the audit and confirm ERROR count is 0 and the only remaining items are the flag-only findings you reported:
bash .claude/skills/doc-audit/scripts/audit-docs.sh --format prettyThen tell the user to review git diff before committing.
Key Rules
- Detection is deterministic and read-only; all edits are yours, applied surgically.
- Truth = SKILL.md frontmatter + manifests. Never invent a command not backed by a SKILL.md.
- Auto-fix only the codes listed in Step 3; everything else is flag-only.
- Leave existing hand-written documentation intact — add, don't rewrite.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

