grill-me
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of instructional text and evaluation data. It does not include any executable scripts, binary files, or external code dependencies.
- [SAFE]: No network operations or external data exfiltration patterns were detected. All interactions are confined to the conversational context between the user and the agent.
- [SAFE]: No credentials, API keys, or sensitive file paths are hardcoded or accessed by the skill instructions.
- [PROMPT_INJECTION]: The skill instructions define a specific persona ('relentless Socratic interviewer') and set clear rules for engagement (e.g., one question at a time). These instructions are constructive and do not attempt to bypass safety filters or override core agent guidelines.
- [DATA_EXPOSURE_AND_EXFILTRATION]: While the skill is designed to process user-provided technical documents (PRDs, architecture plans), it does not contain any mechanisms to transmit this data to third-party services.
Audit Metadata