grill-me
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided designs and instructs the agent to explore the local codebase to answer questions. This creates an attack surface for indirect prompt injection where instructions hidden within the project files or the user's design could influence the agent's behavior.
- Ingestion points: User-supplied plans/designs and local codebase files processed during the interview.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill instructions.
- Capability inventory: The agent is authorized to perform codebase exploration (file reads) to resolve dependencies.
- Sanitization: The skill does not define any input validation or sanitization for the content it processes.
Audit Metadata