agentleFS
gwpicardgwpicard/ai-loop-kit/.agents/maintainer-skills/review-public-docs/SKILL.md

review-public-docs

Skill1 starsChanged 23 days agoMITView on GitHub

Review the shipped prose and the release notes draft before a release. Load by path.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    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
  1. Very specific:

    Run before preparing a release.

    line 8
  2. Distinctive:

    Every part has had one fresh reviewer, every confirmed finding is on the list, and the maintainer has fixed or accepted each one before the …

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review public docs

Run before preparing a release. The shipped prose is README.md, docs/DESIGN.md, CONTRIBUTING.md, SECURITY.md and every Markdown file under kit/, plus the version's notes draft in docs/release-notes/. Split it into at most three parts, and include the relevant source files with each part.

Give each part to a fresh reviewer that has not seen another review. Ask it to check claims in this order: code and checks, release notes, then the docs. Look for stale terms, contradictions, wrong counts, broken references and unsupported claims, such as an old version number or a feature that has changed. A reviewer must confirm each finding in the code or a check before reporting it. Otherwise it reports the concern as unconfirmed.

Collect the reports in .agents/tmp/public-docs-review/ and combine confirmed findings into one list for the maintainer. A release waits until each finding is fixed or the maintainer accepts it. Keep accepted findings and their reasons with the review record.

This review checks that claims are true. The version's notes file also gets a review against the release-notes standard, before the release pull request. .agents/maintainer-skills/release-notes/SKILL.md says how. The quality of the writing, rather than its truth, is reviewed against .agents/maintainer-skills/writing-review/SKILL.md.

The printout

One line for each finding, most serious first: file:line | kind (stale, contradiction, count, reference, unsupported) | what the code or check shows, with its file:line | the fix in one line | confirmed or unconfirmed.

Under the list, give the count of each kind and the findings the maintainer already accepted, with the reason. This review reports and changes nothing.

Done when

Every part has had one fresh reviewer, every confirmed finding is on the list, and the maintainer has fixed or accepted each one before the release is prepared.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.