agentsclimarketplace

Docs audit

Skill stark-ai-de/agent-skills/incubator/skills/repo-maintenance/docs-audit

Public Agent Skills for Codex operations, Cursor operations, Claude operations, repo maintenance, skill maintenance, productivity, and engineering workflows.

Install
npx -y skills add stark-ai-de/agent-skills --skill docs-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Audit repository documentation, README, examples, installation steps, onboarding flow, API docs, and contribution docs. Use when the user asks whether docs are clear, complete, current, or ready for a public repo.

The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.7 KB, as published. Nobody here has run it

Docs Audit

Goal

Find documentation gaps that block users, contributors, maintainers, or agents from understanding, installing, validating, and safely changing the repository.

When to use

  • The user asks whether docs are clear, complete, current, or public-ready.
  • README, examples, onboarding, API docs, or contribution docs need review.
  • Documentation promises may have drifted from actual code and commands.

When not to use

  • The user asks for a full repo maintenance audit; use repo-health-audit.
  • The user wants release notes; use release-manager.
  • The user wants prose editing only and no repo evidence check.

Inputs to inspect

  • README, docs, examples, changelog, contributing guide, security policy, and license.
  • Package scripts, CLI help, config examples, and source files that docs mention.
  • Internal links, install commands, screenshots, and generated docs.

Review rubric

Check audience fit, first-run path, command accuracy, link integrity, examples, maintenance docs, public safety, and agent usability. Read references/docs-rubric.md for details.

Workflow

  1. Identify the intended audiences and main user journeys.
  2. Compare README promises with actual files and commands.
  3. Verify install, validate, and usage examples where practical.
  4. Check internal links and stale references.
  5. Separate user-facing docs gaps from maintainer-only improvements.
  6. Produce a prioritized docs gap list.

Safety rules

  • Do not publish docs, rewrite policy, or change public support commitments without approval.
  • Do not include private URLs, internal hostnames, customer details, or credentials in examples.
  • Mark commands as unverified when they were not run.

References

Read only when needed:

  • references/docs-rubric.md

Scripts

No bundled scripts.

Output format

Return:

  1. Summary
  2. Blocking docs gaps
  3. Confusing or stale sections
  4. Missing examples
  5. Suggested rewrite or new docs
  6. Recommended next action

Failure modes

  • If commands cannot run, report them as unverified instead of broken.
  • If the product goal is unclear, infer it from README and mark the inference.
  • If docs are generated, identify the source file before editing output.

Completion criteria

  • Findings reference docs paths and, when relevant, source-of-truth code.
  • Recommended changes are prioritized by user impact.
  • The next docs action is specific.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.