crisis-holding
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of instructional text for generating crisis communication drafts and does not utilize any tools or execute code.
- [SAFE]: It includes comprehensive safety logic, such as a 'Legal-counsel gate' that halts generation for high-risk incident types (e.g., data security, product safety) unless counsel is involved.
- [SAFE]: The instructions prioritize factual accuracy and explicitly prohibit the AI from speculating or inventing details not provided in the user's initial intake.
Audit Metadata