agentleFS
KSEGITKSEGIT/Version-Sentinel/.agents/skills/audit-dependencies/SKILL.md

audit-dependencies

Skill6 starsChanged 2 days agoMITView on GitHub

Use when the user requests a Version Sentinel dependency audit or sees "Version Sentinel startup audit failed", missing or expired audit evidence, or dependency work blocked by the startup audit.

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 3

    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
  1. Very specific:

    Run the online audit for the user's current project and report the result.

    line 8
  2. Distinctive:

    Use the project directory named in the startup warning or the user's request.

    line 12
  3. Distinctive:

    Run through Bash, quoting both paths:

    line 14
  4. Distinctive:

    Do not disable hooks, bypass verification, edit evidence by hand, or change dependencies to make the audit pass.

    line 28

“Common” is measured: kept by at least 6 in 10 of the 6 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.

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

Audit dependencies

Run the online audit for the user's current project and report the result. The audit records evidence in .version-sentinel/audit.json; it does not install packages or change dependency versions.

Run the audit

  1. Use the project directory named in the startup warning or the user's request. Keep that directory as the command's working directory; do not audit the plugin installation.
  2. Locate the script bundled with this skill. In a plugin install it is ../../scripts/check-versions.sh relative to this skill's directory. In the repository's .agents/skills/ copy it is ../../../scripts/check-versions.sh. Resolve the script to an absolute path. Do not select an older ./scripts/check-versions.sh just because it exists in the user's project, or assume plugin environment variables are set in the terminal.
  3. Run through Bash, quoting both paths:

    cd "<project-directory>" && bash "<absolute-plugin-root>/scripts/check-versions.sh"

    Inside Codex, the script reads CODEX_THREAD_ID automatically. Keep the current session identity; do not invent one or copy audit evidence from another project.

Report the result

  • Report the exit status and the specific diagnostic. DRIFT, UNPINNED, and intentional-pin rows alone do not fail the audit or authorize upgrades.
  • Report completion only when the command exits 0 and prints mandatory online dependency audit completed. A silent exit 0 can mean the plugin is disabled; report that the audit was skipped, without changing the disable setting or claiming current evidence. Separate package-version recording is still required before dependency changes.
  • On failure, keep dependency work blocked. Report registry or prerequisite failures with the retry command. For parser limitations, preserve manifests and npm overrides and report the tool limitation. Missing or invalid manifests need the user's approval before repair or recreation.

Do not disable hooks, bypass verification, edit evidence by hand, or change dependencies to make the audit pass. After the reported cause is resolved, rerun this same command.

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.