- Home
- Skills
- DevOps & Cloud
- Supply Chain Security Assessment
More screenshots
Works with the AI Agents you already use
Supply Chain Security Assessment
Assess your own supply chain end to end: SBOM as inventory, reachability triage, CI/CD gates in order, container hardening, and an incident playbook, with every finding scored and owned.
$5
Supply Chain Security Assessment
Example session with this skill installed
Assess our supply chain before we ship v2.0. Small SaaS: one Node.js app, GitHub Actions CI, Docker deploy to a cloud host, roughly 180 npm dependencies. Here are the workflow files, lockfile, and Dockerfile.
- Read your context and instructions
- Compiled the supply chain security
SCOPE — One Node.js SaaS app, GitHub Actions CI, Docker deploy. Inputs seen: workflow files, lockfile, Dockerfile. No SBOM existed — generated from the production build as finding zero: 187 components vs 162 in the manifest.
THREAT RANKING — Maintainer compromise and stale components first (187 transitive dependencies), CI injection second (one shared workflow, 9 contributors).
FINDINGS (rubric-rated):
- S-01 Critical, owner: release lead. Release workflow token carries broad write permissions and the image promotes unsigned. Impact: anyone who lands a workflow change can ship any artifact as a release. Fix: scope the token per job, sign the image, verify at admission. Due: 48 hours.
- S-02 Critical. Reachable, publicly exploited flaw in a transitive file-upload parser. Fix: upgrade, retest by SBOM diff. Due: 48 hours.
- S-03 High. Five third-party actions pinned to floating major-version tags. Fix: pin to full commit hashes. Due: 7 days.
- S-04 High. Reachable flaw in the session library on the login path. Fix: upgrade. Due: 7 days.
- S-05 Medium. Container runs as root from a full base image on a floating tag; the OS layer was never scanned. Fix: pin by digest, minimal base, non-root user, scan all layers. Due: 30 days.
Backlog: 41 scanner findings not reachable on current call paths — review date set, re-triage on the next SBOM diff.
DECISION — Ship is blocked on S-01 and S-02 only. Everything else carries an owner and a date inside its rubric window, and the incident playbook now names who scopes, contains, and rebuilds for this project.
Connects securely to your tools. The creator never sees your data.
About this skill
A defensive assessment method for the supply chains you own: your dependencies, your CI/CD pipeline, and your containers. It starts by ranking five attack vectors — dependency confusion and typosquatting, maintainer compromise, build tampering, CI injection, and stale components — against the actual shape of your project, so the audit spends time where your risk really is. The SBOM becomes the inventory of record: generated from the build rather than the manifest, covering transitive dependencies, and diffed in review so a dependency change nobody reviewed cannot slip through. Scanner output is triaged by reachability, exploitability in your context, and known exploitation instead of raw count or base score, which is how reachable mediums stop hiding behind unreachable criticals. The pipeline itself is gated in order — secrets, dependencies, build integrity with signing and provenance, and admission verification — and assessed before dependency findings are ranked, because one over-permissioned workflow compromises every artifact it produces. Containers are hardened as a ladder: minimal pinned base images, per-layer scanning, signing, and restrictive runtime policy. Every finding is filed on a standard template, rated on a four-row scoring rubric with fixed response windows, and given an owner and a retest step. A worked example runs a full pre-launch assessment on a small web service, and the closing incident playbook — scope, contain, rebuild, verify, harden — is written before you need it.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
- It directs and structures the assessment; it does not run your scanners or read your pipeline directly — you supply the workflow files, lockfiles, and scanner output.
- Reachability calls are only as good as the inventory and call-path information you provide; an incomplete SBOM caps the accuracy of the triage.
- Rubric ratings are calibrated judgment for prioritization, not a compliance certification; regulated environments may also need a formal audit.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 24 days ago
Permissions
No special permissions declared or detected