security-reviewer
Use when explicitly asked to run the security-reviewer subagent or when another skill requires the security-reviewer agent card.
When & Why to Use This Skill
The security-reviewer is a specialized Claude subagent designed to perform rigorous security audits and vulnerability assessments. It streamlines the DevSecOps workflow by identifying potential security flaws, hardcoded secrets, and compliance risks, ensuring that code changes adhere to industry-standard security best practices before they are merged.
Use Cases
- Automated Vulnerability Scanning: Use the subagent to scan pull requests for common security threats such as SQL injection, Cross-Site Scripting (XSS), and insecure data handling.
- Security-Focused Code Audits: Explicitly trigger the reviewer to perform a deep dive into sensitive modules, such as authentication logic or encryption implementations, to find logic flaws.
- Compliance and Best Practice Verification: Ensure that new code follows internal security guidelines and architectural constraints by providing specific scope and context to the subagent.
- Risk Assessment for PRs: Integrate the security-reviewer into the main development workflow to provide a dedicated security perspective on complex code changes.
| name | security-reviewer |
|---|---|
| description | Use when explicitly asked to run the security-reviewer subagent or when another skill requires the security-reviewer agent card. |
Security Reviewer Subagent
Use the security-reviewer agent card to handle this specialized task.
Run
codex-subagent security-reviewer <<'EOF'
[Provide the task context, scope, and any issue/PR numbers.]
EOF
Notes
- Include concrete scope and constraints in the context block.
- Fold the subagent output back into the main workflow.