agentleFS
thinkyou0714thinkyou0714/github-flow-kit/gh-repo-security-audit/SKILL.md

gh-repo-security-audit

Skill0 starsChanged 5 months agoMITView on GitHub

OpenSSF-aligned security posture audit across all repos in a GitHub account: default workflow token permissions, allowed-actions policy, branch protection, secret scanning + push protection, and Dependabot alerts. Reports WARN (fixable gaps) vs INFO (opinionated hardening). Read-only by default; the only optional mutation is enabling Dependabot alerts. Use when you ask: "repo security audit", "OpenSSF audit", "are Dependabot alerts on?", "GitHub hardening check", "repo セキュリティ監査", "Actions セキュリティ横断". DO NOT USE FOR: writing code, changing branch protection automatically, or the PR-approval toggle (use gh-pr-perm-audit for that).

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 gh auth status — if not authenticated, output ⚠️ gh CLI not authenticated.

    line 19
  2. Distinctive:

    no branch protection on the default branch

    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

GitHub Repo Security Posture Audit (OpenSSF-aligned)

Audit every repo in an account against OpenSSF Scorecard-style checks and report a clear posture. Read-only by default. The single optional mutation is enabling Dependabot alerts (a pure security gain). Everything else is reported as a human-gated recommendation. See references/openssf-checks.md for per-check rationale + sources.

Setup Check

Run gh auth status — if not authenticated, output ⚠️ gh CLI not authenticated. Run: gh auth login and stop.

Step 1: Resolve scope

  • --owner <name> or default OWNER=$(gh api user --jq .login).
  • Repos: gh repo list "$OWNER" --no-archived --limit 200 --json name --jq '.[].name'.

Step 2: Per-repo checks (GET-only)

For each repo, gather in as few calls as possible:

# one snapshot of the repo object: visibility, default branch, secret scanning
gh api "repos/$OWNER/$REPO" \
  --jq '[.visibility, .default_branch, (.security_and_analysis.secret_scanning.status // "n/a")] | @tsv'
gh api "repos/$OWNER/$REPO/actions/permissions/workflow" --jq '.default_workflow_permissions'   # read|write
gh api "repos/$OWNER/$REPO/actions/permissions"          --jq '.allowed_actions // "all"'         # all|local_only|selected
gh api "repos/$OWNER/$REPO/branches/$DEFAULT_BRANCH/protection" >/dev/null 2>&1 && echo yes || echo none
gh api "repos/$OWNER/$REPO/vulnerability-alerts"          >/dev/null 2>&1 && echo on  || echo off

Step 3: Severity model

  • WARN (fixable gap):
    • Dependabot alerts off
    • default_workflow_permissions = write (least-privilege violation; should be read)
    • a public repo with secret scanning not enabled
  • INFO (opinionated hardening — not a defect):
    • no branch protection on the default branch
    • allowed_actions = all

Step 4: Report

# Repo Security Audit — <owner> — <DATE>

> Generated by github-flow-kit/gh-repo-security-audit (OpenSSF-aligned)

| Repo | vis | def perms | allowed | branch prot | secret scan | Dependabot |
|---|---|---|---|---|---|---|
| <repo> | public | read | all | none | enabled | on |

## ⚠️ WARN (fixable): <N>
- <repo>: Dependabot alerts OFF
## ℹ️ INFO (hardening): <N>
- <repo>: no branch protection / allowed_actions=all

Step 5: Remediation

Safe, optional mutation — enable Dependabot alerts (only with --enable-dependabot; pure gain, reversible):

gh api -X PUT repos/<owner>/<repo>/vulnerability-alerts     # enable  (DELETE to undo)

Human-gated (never auto-applied) — print commands, let the user decide:

  • Token-Permissions: keep default read; declare per-workflow permissions: blocks in YAML.
  • Branch protection (esp. public + bot-PR repos): require reviews from someone other than the actor.
  • allowed_actions → selected + verified creators (stage rollout; third-party actions may break).
  • SHA-pin actions (uses: owner/action@<40-hex-sha>): add dependabot.yml github-actions ecosystem, or use ratchet/pin-github-action.
  • Review any pull_request_target / workflow_run workflows for untrusted PR-head checkout and ${{ github.event.* }} injection into run:.

Summary Output

✅ Repo Security Audit — <N> repos

  ⚠️ WARN (fixable) : <n>   (mostly Dependabot OFF — fix with --enable-dependabot)
  ℹ️ INFO (hardening): <n>
  ✅ secure          : <n>

📋 Next: --enable-dependabot for the safe fix; branch protection / allowed_actions / SHA-pin stay manual.

⭐ Found this useful? Star → https://github.com/thinkyou0714/github-flow-kit

Error Handling

CodeConditionExit
RS-001gh not authenticated2
RS-002owner has no repos0
RS-003per-repo API error (no admin / archived)skip that repo
RS-004API rate limitedstop, show reset time

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.