clean-ai-slop
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on code files that may have been generated by AI or obtained from external sources, creating a potential surface for instructions embedded in data.\n
- Ingestion points: The skill explicitly targets files generated or modified by AI as defined in the 'Hard Gates' and 'When To Use' sections.\n
- Boundary markers: There are no specific delimiters or markers (e.g., XML tags or block markers) defined to separate the target code from the agent's instructions.\n
- Capability inventory: The skill requires the agent to read, modify, and potentially execute code via test suites and linting tools.\n
- Sanitization: The risk is mitigated by a strict multi-pass process that requires existing and regression tests to pass after every single modification, and by using external tooling (compilers, linters) to identify dead code.
Audit Metadata