ai-slop-cleaner
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by analyzing and editing external code files, which creates a potential surface for indirect prompt injection. Malicious instructions could be embedded in code comments or data within the files being processed, potentially influencing the agent's behavior during the cleanup process.
- Ingestion points: Code files identified in the feature area or provided as a file list scope (SKILL.md).
- Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" warnings for the content being processed.
- Capability inventory: The skill performs file modifications, runs regression tests, and executes quality gate tools like linters and typecheckers (SKILL.md).
- Sanitization: No content sanitization or validation of embedded instructions within the source files is mentioned.
Audit Metadata