performance-testing-review-ai-review
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as an instructional guide and implementation playbook for setting up code review workflows. It utilizes legitimate security and quality tools such as SonarQube, CodeQL, and Semgrep.
- [COMMAND_EXECUTION]: The Python and shell examples demonstrate the use of subprocesses and CLI commands to orchestrate analysis tools. These are standard practices for the described DevOps automation tasks.
- [DATA_EXFILTRATION]: Integration examples follow security best practices by using environment variables (e.g., GITHUB_TOKEN, OPENAI_API_KEY) and GitHub Secrets for authentication, rather than hardcoding credentials.
- [PROMPT_INJECTION]: The skill exhibits an Indirect Prompt Injection attack surface, which is a common design consideration for code review agents.
- Ingestion points: The system processes untrusted code diffs and PR descriptions (via $ARGUMENTS and script variables).
- Boundary markers: The examples use markdown headers (e.g., Modified Code:) but do not include explicit instructions for the AI to ignore embedded commands within the diffs.
- Capability inventory: The skill's demonstrated capabilities include running static analysis tools and posting comments back to pull requests.
- Sanitization: The provided code templates do not demonstrate input sanitization or escaping for the interpolated content. Users implementing this blueprint should include robust delimiters and sanitization logic.
Audit Metadata