code-review
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run maintenance and verification scripts within the repository, such as
npm run check-pr-approvalsandscripts/Update_CheatSheets_Index.py. These are legitimate developer operations within the official OWASP project environment. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted input from pull requests and external URLs. The instructions mitigate this risk by requiring the agent to manually fetch and verify all sources, use rigorous review dimensions (Security, Practicality, Sources), and explicitly check for unverified AI-generated content or 'link spam'.
- Ingestion points: Pull request diffs, full files in
cheatsheets/orcheatsheets_draft/, and external URL content. - Boundary markers: None explicitly defined in the prompt structure.
- Capability inventory: File system read access, network access (to verify links), and command execution via npm.
- Sanitization: Relies on the agent's 'Security correctness' review phase to manually identify and report malicious or incorrect advice.
Audit Metadata