skills/freedomintelligence/openclaw-medical-skills/crisis-detection-intervention-ai/Gen Agent Trust Hub
crisis-detection-intervention-ai
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements mental health crisis detection through documented implementation patterns and utility scripts that prioritize safety, human intervention, and immediate access to professional resources.
- [SAFE]: External network references, such as the SAMHSA government API and HuggingFace model registries, are well-known technology and public health services appropriate for the skill's functionality.
- [SAFE]: Data management guidelines within the documentation promote high security standards for handling sensitive user content, including mandatory encryption at rest and transit, access auditing, and short-term data retention.
- [SAFE]: Included scripts (scripts/crisis_detector.ts and scripts/model_evaluator.ts) perform logic-based text analysis using regex patterns and do not involve suspicious command execution or unauthorized data exfiltration.
- [SAFE]: Prompt templates shown in implementation examples utilize delimiters to help isolate user-provided content from agent instructions, mitigating risks of indirect prompt injection.
Audit Metadata