polish
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external code and tests, which presents a surface for potential indirect prompt injection.
- Ingestion points: Data ingested via the
$codeand$testsplaceholders inSKILL.md. - Boundary markers: None defined to separate instructions from the content being processed.
- Capability inventory: The skill identifies code violations, applies fixes to files, and executes verification checks.
- Sanitization: No specific filtering or escaping mechanisms for the input code are mentioned.
Audit Metadata