crisis-response-protocol

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a robust framework for crisis detection and intervention based on established mental health best practices.
  • [SAFE]: All external resource links point to reputable and verified emergency services, including the 988 Suicide & Crisis Lifeline and SAMHSA.
  • [SAFE]: The implementation patterns include safety-first measures such as human-in-the-loop review triggers and automated response validation to prevent the AI from providing unsafe or minimizing advice.
  • [SAFE]: No malicious patterns, prompt injections, obfuscation, or unauthorized data exfiltration mechanisms were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 08:55 PM