- Home
- Skills
- Agents & Orchestration
- CI CD Pipeline Hardening Agent
Works with the AI Agents you already use
CI CD Pipeline Hardening Agent
Harden CI/CD pipelines against supply chain attacks via SHA-256 pinning and least-privilege OIDC configuration.
$15
CI CD Pipeline Hardening Agent
Example session with this skill installed
Harden my github actions workflow in .github/workflows/deploy.yml to prevent supply chain attacks and fix token permissions.
- Read your context and instructions
- Compiled the ci cd pipeline
I have audited .github/workflows/deploy.yml. I replaced mutable tags with SHA-256 hashes and restricted the GITHUB_TOKEN to the minimum required permissions.
Hardened Workflow Configuration
name: Production Deployment
on:
push:
branches: [main]
permissions:
contents: read
id-token: write # Required for OIDC auth
jobs:
deploy:
runs-on: ubuntu-latest
steps:
- name: Checkout
# Using SHA-256 for actions/checkout@v4.1.1
uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11
- name: Configure AWS Credentials
# Using SHA-256 for aws-actions/configure-aws-credentials@v4.0.2
uses: aws-actions/configure-aws-credentials@010d0da01d0b5a38af11e9c3470dbfad0f28c927
with:
role-to-assume: ${{ secrets.AWS_ROLE_ARN }}
aws-region: us-east-1
- name: Deploy
run: |
echo "Deploying to production environment..."
# Script execution is wrapped to prevent injection
./scripts/deploy.sh
Next steps
1.
Verify Hashes: Use git ls-remote to verify the SHA-256 hashes correspond to the trusted versions of the actions.
2.
Service Provider Trust: Ensure your cloud provider (AWS/Azure/GCP) is configured to trust the OIDC subject claim from this specific repository.
3.
Commit and Test: Push these changes to a feature branch to ensure the id-token permission is sufficient for your deployment role.
Connects securely to your tools. The creator never sees your data.
About this skill
Broken CI/CD pipelines are the primary vector for supply chain attacks. This skill identifies and fixes critical vulnerabilities in your GitHub Actions, GitLab CI, or Jenkins configurations before they are exploited.
What it does
- Pin Dependencies: Replaces mutable version tags with immutable SHA-256 commit hashes to prevent tag-switching attacks.
- Enforce Least Privilege: Configures minimal OIDC
permissionsblocks to limit token scope toid-tokenandcontents: read. - Inject Security Headers: Audits and applies secret masking and environment isolation settings.
- Workflow Auditing: Scans YAML files for shell injection vulnerabilities and unsafe runner configurations.
How it works
- Source Audit: The agent scans
.github/workflows/,.gitlab-ci.yml, or Jenkinsfiles in your current workspace. - Hardening Analysis: It identifies unpinned actions, excessive token permissions, and script injection risks.
- Remediation: The agent generates or modifies the YAML code to apply cryptographic immutability and OIDC best practices.
- Verification: It references architectural patterns to ensure the new configuration meets DevSecOps compliance standards.
Frameworks & tools
This skill supports GitHub Actions, GitLab CI/CD, and Jenkins. It utilizes SHA-256 pinning, OIDC token exchange, and local runner isolation.
Why this beats prompting it yourself
Generic LLMs often suggest version tags like @v4, which are mutable and insecure. This skill enforces SHA-256 pinning by default and applies precise OIDC permission blocks that developers often forget or misconfigure.
Use cases
- Convert a GitHub Action using version tags to full SHA-256 immutability.
- Secure a pipeline vulnerable to secret exfiltration via malicious PRs.
- Implement least-privilege OIDC tokens for AWS or Azure deployments.
- Audit a legacy Jenkinsfile for unsafe shell script execution.
Known limitations
Analysis is restricted to the local workspace. It cannot access system-level runner configurations or external secret managers beyond the pipeline definitions.
Restricted to workspace files. Cannot modify system-level runner settings or external secret manager dashboards.
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 any payment.
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 4 days ago
Permissions
Target Use Cases: - Harden our GitHub Actions workflow against supply chain attacks. - Audit GitLab CI pipeline for secret exfiltration risks. - Pin all GitHub action versions to full SHA-256 commits.