grilling
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for a structured interaction model ('grilling') to explore goals and technical decisions. It uses logical filtering ('Altitude') to manage user communication and decision-making.
- [INDIRECT_PROMPT_INJECTION]: The skill has the capability to ingest information from the environment (filesystem, tools) by dispatching sub-agents to find facts. While this creates a surface for processing potentially untrusted data from the codebase, it is a standard functional requirement for development-oriented agents and lacks any patterns that would exploit this access for malicious purposes. Evidence includes the instruction: 'When a frontier question needs a fact from the environment (filesystem, tools, etc.), dispatch a sub-agent to find it'.
Audit Metadata