agentleFS
mtarcuremtarcure/claude-vibe-squad/.agents/skills/supply-chain-audit/SKILL.md

supply-chain-audit

Skill220 starsChanged 20 days agoMITView on GitHub

Use when auditing whether anything outside the project can silently alter a build or published release—inventory dependencies, base images, binaries, generators, CI actions, install hooks, secret exposure, and publisher controls; verify pins, digests or signatures, reproducibility, and one artifact's end-to-end provenance. Use dependency-health-triage for advisory reachability alone; this owns build and release integrity.

9 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

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Check pinning and integrity for each input — exact versions, lockfile presence, checksum or digest verification, and signature verification …

    line 13
  2. Distinctive:

    Check third-party CI actions specifically — pinned to a commit SHA rather than a mutable tag, and reviewed for the permissions they request.

    line 15
  3. Distinctive:

    Look for typosquat and confusion risk: internal package names resolvable from a public registry, recently-renamed or transferred packages, …

    line 16
  4. Distinctive:

    Verify the release path: who can publish, whether publishing requires review, whether artifacts are signed, and whether the published …

    line 17
  5. Distinctive:

    Check install-time execution — post-install scripts, build hooks, and container entrypoint fetches — as these run with developer or build …

    line 18
  6. Distinctive:

    All build inputs are inventoried with their pinning and integrity-verification state.

    line 23

“Common” is measured: kept by at least 6 in 10 of the 8 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

Supply Chain Audit

Audit everything that enters a build or a release without being written by the project, and everything that can modify the release on its way out.

Steps

  1. Inventory inputs: package dependencies, base images, downloaded binaries and installers, CI actions and plugins, and build-time code generators.
  2. Check pinning and integrity for each input — exact versions, lockfile presence, checksum or digest verification, and signature verification where the ecosystem supports it. A floating tag is an unpinned input.
  3. Audit the CI/CD pipeline as attacker surface: who can trigger a build, what secrets each job can read, whether pull-request builds from forks get privileged tokens, and whether build steps can be influenced by the code they are building.
  4. Check third-party CI actions specifically — pinned to a commit SHA rather than a mutable tag, and reviewed for the permissions they request.
  5. Look for typosquat and confusion risk: internal package names resolvable from a public registry, recently-renamed or transferred packages, and single-maintainer packages with sudden ownership changes.
  6. Verify the release path: who can publish, whether publishing requires review, whether artifacts are signed, and whether the published artifact can be reproduced from the tagged source.
  7. Check install-time execution — post-install scripts, build hooks, and container entrypoint fetches — as these run with developer or build privilege.
  8. Trace the provenance chain end to end for at least one artifact, and record every point where the chain relies on trust rather than verification.
  9. Rank findings by whether they permit silent modification of shipped code; that class outranks everything else here.

Acceptance

  • All build inputs are inventoried with their pinning and integrity-verification state.
  • CI permissions, fork-build behavior, and third-party action pinning are checked and recorded.
  • Dependency-confusion and ownership-change risks are examined for internal names.
  • The release path is traced end to end, with every trust-not-verification point named.
  • Findings that permit silent modification of shipped artifacts are ranked highest.

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.