grill-me
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional text designed to guide the agent's conversational logic and persona.\n- [SAFE]: The instructions for codebase exploration utilize standard, pre-approved agent tools to gather context for the interview process.\n- [SAFE]: No external dependencies, remote scripts, persistence mechanisms, or credential access patterns were detected.\n- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes untrusted user-provided designs and project plans.\n
- Ingestion points: User-supplied plan or design descriptions via the conversation (SKILL.md)\n
- Boundary markers: None present\n
- Capability inventory: Tools for codebase exploration and standard text processing\n
- Sanitization: None identified
Audit Metadata