CI/CD Pipeline Security Audit
Reads your build and deploy pipelines for the ways a stranger's pull request or a poisoned dependency could reach production.
What it examines
- Pull requests that run with your secrets
- Text from a pull request pasted into a shell command
- Long-lived cloud keys where short-lived ones would do
- Actions and scripts that can change without a review
- Runners that keep state between jobs
- Production deploys with no approval
What you get back
A findings file. Each item has the evidence, how serious it is, and a fix.
Dependencies
- An AI coding agent that reads SKILL.md files, such as Claude Code, Codex or Gemini CLI
- Read access to every pipeline definition: GitHub Actions, GitLab CI, Bitbucket Pipelines, Azure Pipelines, Jenkinsfiles and the scripts they call
- Optional: an export or screenshots of branch protection, required reviews, environment rules and runner settings. These live outside the repository.
- Optional: zizmor, actionlint or OpenSSF Scorecard, if already installed. Their output is treated as leads only.
- Python 3 with the jsonschema package, to check findings.json before fixing anything
How it works
The audit only reads. It never triggers a workflow, pushes a branch or reads a secret value. It reports where a secret lives and how it reaches a job, not what it contains.
It finds the pipeline systems in use afresh and runs every check against whichever platform it finds.
Settings held outside the repository, such as branch protection, are recorded as insufficient evidence unless you supply an export.
You get findings.json and findings.md. The remediate skill reads findings.json and tightens the pipeline files.
Setup
- Unzip the download, then copy the audit and remediate folders into your agent's skills folder (for Claude Code, .claude/skills/) in the repository you want to check.
- Export your branch protection and environment settings if you can. The audit is stronger with them.
- Open the repository in your agent and ask it to run the i3-cicd-pipeline-security-audit skill.
- Check the result with the validator: pip install jsonschema, then python3 schema/validate.py findings.json.
- Ask your agent to run the i3-cicd-pipeline-security-remediate skill using findings.json.
Run it with your agent
Copy the prompt, fill in the parts in square brackets, and paste it into your agent.
Run the i3-cicd-pipeline-security-audit skill on this repository. Branch protection and environment settings export: [path, or none]. Never trigger a workflow or read a secret value. Mode: audit only. Do not change any code. Write findings.json and findings.md in the repository root.
Run the i3-cicd-pipeline-security-remediate skill using findings.json. Work through the findings I choose, and leave the rest open.