AgentStack
SKILL verified Apache-2.0 Self-run

Github Actions Hardening

skill-ravisingh11-agent-safe-engineering-github-actions-hardening · by ravisingh11

Review and harden GitHub Actions workflows for supply-chain risk, token permissions, secret handling, script injection, OIDC, reusable workflow access, and CI governance. Use when asked to audit CI/CD security, fix workflow risk, or standardize Actions across repositories.

No reviews yet
0 installs
2 views
0.0% view→install

Install

$ agentstack add skill-ravisingh11-agent-safe-engineering-github-actions-hardening

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Github Actions Hardening? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

GitHub Actions Hardening

Use this skill when reviewing or changing GitHub Actions workflows.

Default policy: treat CI as an explicit repository decision. Do not add workflow files, required GitHub Actions checks, or hosted runner dependencies unless the repository owner has selected GitHub Actions for that repo.

Workflow

  1. Inventory .github/workflows, local composite actions, reusable workflows, required checks, Dependabot config, and repository Actions settings.
  2. Identify trust boundaries: pull requests, forks, release jobs, deployment jobs, secrets, cloud credentials, and third-party actions.
  3. Check for least-privilege permissions, unsafe inline scripts using untrusted context, plaintext or transformed secrets, mutable third-party action references, and long-lived cloud credentials.
  4. Prefer OIDC for cloud access when the provider and deployment model support it.
  5. Record concrete risks and fix the smallest safe surface first.

Decision Rules

  • Use permissions: contents: read as the default baseline; elevate per job only when required.
  • Prefer deleting obsolete workflow files over keeping inert CI definitions.
  • Treat pull_request_target, deployment secrets, and write tokens as high-risk until proven safe.
  • Prefer full-SHA pinning for third-party actions in sensitive workflows; document any tag-based exception.
  • Do not move every workflow to reusable workflows by default. Use reusable workflows only when the repo set benefits from central maintenance.
  • For reusable workflows in a central repository, confirm repository Actions access policy before expecting downstream repos to call them.

Verification

  • Run workflow syntax validation where tooling exists.
  • Re-read changed workflows for untrusted context interpolation.
  • Confirm required checks and branch rules still reference valid workflow names.
  • For deploy jobs, confirm credentials are scoped and not printed to logs.

References

  • GitHub Actions security baseline: references/security-baseline.md

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.