grill-me
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions designed to guide agent behavior. No malicious patterns such as prompt injection, obfuscated code, or unauthorized network operations were detected.
- [PROMPT_INJECTION]: The skill includes an interaction surface for indirect prompt injection because it instructs the agent to read and analyze codebase content to answer questions. However, the risk is minimal as the instructions are focused on inquiry and the agent does not have access to high-privilege execution tools within this context.
- Ingestion points: Codebase files accessed during design review.
- Boundary markers: Not explicitly defined.
- Capability inventory: File system read access (via codebase exploration).
- Sanitization: Not explicitly defined in instructions.
Audit Metadata