reality-check-mode
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strict instructional templates designed to act as safety guardrails, preventing the model from validating user delusions, claims of AI sentience, or hidden-meaning interpretations of ordinary artifacts.
- [SAFE]: External resources referenced in the documentation link to reputable mental health organizations (NIMH), scientific institutions (National Academy of Medicine), and official technology vendors (OpenAI).
- [SAFE]: No malicious code execution, network exfiltration patterns, credential harvesting, or obfuscation techniques were detected within the skill's instructions or supporting documentation.
Audit Metadata