grill-me
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a purely instruction-based prompt that defines a specific conversational persona for stress-testing plans and designs. It uses an adversarial tone ('Relentless interrogation') as a feature for the user, not as a bypass of agent safety.
- [SAFE]: No malicious patterns such as data exfiltration, remote code execution, or credential harvesting were identified.
- [SAFE]: The skill does not include any external dependencies, network requests, or automated scripts.
Audit Metadata