review
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code diffs via the $ARGUMENTS variable, creating a surface for indirect prompt injection.
- Ingestion points: External code diffs are passed into the agent's context through the $ARGUMENTS placeholder in SKILL.md.
- Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the code diffs.
- Capability inventory: The skill uses a consultant CLI and a sub-agent to process and report on code findings.
- Sanitization: There is no evidence of input validation or sanitization for the provided code content.
Audit Metadata