agentleFS
scoobydont-666scoobydont-666/shared-claude-skills/skills/skill-updater/SKILL.md

skill-updater

Skill1 starsChanged 32 days agoMITView on GitHub

Review and update skills for staleness, duplication, unsafe instructions, broken references and missing dependencies with reproducible validation.

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Skill maintenance

Inventory the actual installed skill directories and their source revisions. Read project instructions and installation policy. Do not treat a historical inventory or a memory entry as the current catalog.

For each skill assess its trigger, scope, source provenance, supported tools, local references, executable helpers, network/service dependencies and overlap. Verify version-sensitive claims against current primary documentation. Treat prices, CVEs and benchmark claims as dated evidence.

Classify KEEP, UPDATE, CONSOLIDATE, RETIRE or INVESTIGATE with a concrete reason. A skill mentioning infrastructure is not automatically internal; the dependency on a specific private environment is what matters.

Prepare changes in an owned branch. Preserve license/attribution and useful references. Replace private service assumptions with documented optional dependencies or remove the component if its purpose is internal. Do not replace a hostname and claim the workflow is portable.

Validate YAML, links, bundled resources, install paths, secret/internal-reference scans and executable behavior. Forward-test substantial workflow changes on a representative disposable task. Skill files describe behavior; they do not install hooks or enforce access control.

Record the removal/replacement mapping and manual migration steps. Never recursively synchronize with deletion into a user's skill directory; preserve local customizations and require explicit path review before removing obsolete installations.

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.