code-review
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted user code, which creates a surface for indirect prompt injection.
- Ingestion points: User-provided code snippets or files processed during the review workflow.
- Boundary markers: The instructions lack specific delimiters or isolation markers to separate analyzed code from agent instructions.
- Capability inventory: The skill performs file reading to load standards and file writing to save reports to the reviews/ directory.
- Sanitization: No explicit logic is defined to ignore or sanitize instructions embedded in the code under review.
- [SAFE]: The skill instructions, evaluation cases, and security reference files do not contain obfuscated content, hardcoded credentials, unauthorized network calls, or malicious commands.
Audit Metadata