grill-me
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to 'explore the codebase' to answer questions, creating a surface where instructions in codebase files could influence behavior.
- Ingestion points: Local codebase files (SKILL.md).
- Boundary markers: No delimiters or instructions to ignore embedded commands are present.
- Capability inventory: Implicit file reading and interpretation.
- Sanitization: No validation or sanitization of codebase content is mentioned.
Audit Metadata