grilling
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided plans or design documents which could contain malicious embedded instructions.
- Ingestion points: User-provided plans or design documents processed by the agent in SKILL.md.
- Boundary markers: No delimiters or specific "ignore embedded instructions" instructions are present to differentiate user data from agent instructions.
- Capability inventory: The skill grants the agent permission to "explore the codebase" in SKILL.md, indicating file system access capabilities that could be abused if the input plan contains injection attacks.
- Sanitization: No sanitization or validation of the input plan is described.
Audit Metadata