engineering-code-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a specialized persona for performing code reviews. No malicious instructions, obfuscation, or unauthorized access patterns were identified in the SKILL.md or the run.sh script.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external code, which constitutes an ingestion point for untrusted data.
- Ingestion points: Code snippets provided for review through the task argument or stdin in scripts/run.sh.
- Boundary markers: None defined in the instructions to separate untrusted code from agent instructions.
- Capability inventory: No dangerous capabilities (tools, network access, or file system modifications) are requested in the frontmatter or used in the instructions.
- Sanitization: No explicit sanitization or escaping of the input code is performed, but the lack of executable capabilities mitigates the risk.
Audit Metadata