Works with the AI Agents you already use
Secrets Leak Responder — Committed a Key? Ordered Runbook
Scanners find the leak. This is what happens next: rotate-first doctrine, per-provider playbooks, honest history-scrubbing, blast-radius audit.
$12
Secrets Leak Responder — Committed a Key? Ordered Runbook
Example session with this skill installed
Pushed my OpenAI API key to a public GitHub repo 20 minutes ago. Made the repo private already — am I safe?
- Read your context and instructions
- Compiled the secrets leak responder
Secret Leak Response — OpenAI key — severity: EXPOSED (20-min public window = assume scraped; private-now changes nothing retroactively)
FIRST 30 MIN:
- Delete the key in console NOW — rotation before anything.
- Deploy replacement via secret manager.
- Check usage dashboard for foreign spend.
FIRST 2 HRS: 4. GitHub audit log — who forked/pulled. 5. Blast radius: key scope.
TODAY: 6. filter-repo scrub + force-push ⚠ scrub trap: forks/caches/CI keep it anyway. 7. Notify security owner.
THIS WEEK: 8. gitleaks pre-commit + push protection. Least-privilege replacement.
A clean repo with the same live key is still compromised.
Connects securely to your tools. The creator never sees your data.
About this skill
The moment after "I committed a key to git" is not the moment to figure out what to do. This skill is the ordered runbook — built on one rule: rotate before scrub. Deleting the commit doesn't un-leak the key; it was already visible, scraped, forked, or CI-logged. Revocation is the only fix.
What it does
- Severity triage —
SEALED/EXPOSED/COMPROMISED, calibrated to how far the leak actually went. - Timed checklist — first 30 minutes → first 2 hours → today → this week. Every action names its owner.
- Per-provider playbooks — AWS, GitHub, OpenAI/Anthropic, Stripe, GCP, Slack, database creds, SSH keys, JWT secrets (each rotates differently — AWS keeps inactive keys for audit, JWT needs rolling verification or every session dies at once).
- Honest history scrub —
git filter-repo/BFG steps plus the scrub trap: forks, caches, CI logs, and anyone who already pulled keep the key. Scrubbing limits future exposure; it fixes nothing retroactively. - Blast-radius audit — what the key could reach, which usage logs to check (CloudTrail, audit log, dashboards), signs of actual abuse.
- Notification matrix — who to tell, when, why.
COMPROMISEDescalates to full incident mode.
Known limitations
A runbook, not an executor — it doesn't rotate keys or rewrite history for you. Provider console steps are marked verify where they may have drifted. Proactive secret scanning is out of scope — this is response, not detection.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Guidance runbook — it doesn't execute rotations or scrubs for you. Provider console steps marked 'verify' where they may have drifted. Detection (scanning for leaks proactively) is out of scope — this is response, not prevention.
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 10 days ago
Permissions
No special permissions declared or detected