grill-diff
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill is designed to perform security audits on code diffs, including specific instructions for an 'attacker' specialist to check for input validation, secret exposure, and injection risks.
- [SAFE]: The tool retrieves code changes from local files and GitHub PR URLs for analysis and does not involve executing untrusted content or performing unauthorized network operations.
- [PROMPT_INJECTION]: The skill ingests untrusted code diffs and pull request content, which creates a potential indirect prompt injection surface. This is mitigated by the structured review process and the 'attacker' specialist instructions that explicitly look for injection risks. Ingestion points: changed files and pull request URLs. Boundary markers: absent. Capability inventory: read-only access to codebase files, specifications, and tests. Sanitization: none specified.
- [NO_CODE]: The skill consists entirely of instructional markdown files and does not include any accompanying scripts or binary executables.
Audit Metadata