grill-me
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's core functionality provides a structured questioning framework ('grilling') to assist in decision-making and project planning, which is a benign and intended use of agent capabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided plans and ideas which can act as a vector for indirect instructions.
- Ingestion points: The agent ingests user-provided text (plans, designs, or ideas) as the primary input for the interview rounds.
- Capability inventory: The agent has permission to explore the filesystem and tools, and can dispatch sub-agents to gather 'facts' based on the user's plan.
- Boundary markers: There are no explicit delimiters or instructions provided to separate the user-provided content from the agent's internal logic.
- Sanitization: No explicit sanitization or validation of the user's input is performed before the agent uses it to drive environment exploration.
Audit Metadata