grill-me
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions are purely conversational and do not include any commands for data exfiltration, remote code execution, or privilege escalation. The skill explicitly instructs the agent not to claim slash commands have executed unless they actually have, which is a positive safety control.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of user plans and designs.
- Ingestion points: Untrusted data enters the context when the user provides a plan or design to be processed by the interview workflow (SKILL.md).
- Boundary markers: The instructions do not define any delimiters or warnings to ignore instructions embedded within the user's plan.
- Capability inventory: Analysis of the skill reveals no capabilities for subprocess calls, file system writes, or network operations.
- Sanitization: There is no evidence of sanitization or filtering for user-provided content.
Audit Metadata