crisis-response-protocol
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements safety-critical logic for crisis intervention without introducing security risks. It follows best practices by ensuring AI does not act as a sole responder and provides pathways to human support.
- [EXTERNAL_DOWNLOADS]: The skill references established, high-trust organizations including the 988 Suicide & Crisis Lifeline, SAMHSA, and Alcoholics Anonymous. These references are used for safety resource mapping and are appropriately documented.
- [COMMAND_EXECUTION]: No shell command execution or unauthorized subprocess spawning is present in the skill.
- [DATA_EXFILTRATION]: The skill demonstrates safe data handling by explicitly noting that conversation content should not be included in notifications sent to emergency contacts, maintaining privacy while prioritizing safety.
- [PROMPT_INJECTION]: The safety instructions provided for the AI (e.g., 'CRITICAL SAFETY INSTRUCTIONS') are legitimate guidance to ensure responsible behavior and do not attempt to bypass safety filters or override system constraints for malicious purposes.
Audit Metadata