brain-unseal
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists exclusively of Markdown instructions and JSON test cases. It does not include any scripts (Python, Node.js, Shell), binaries, or executable commands, posing no risk of local or remote code execution.
- [SAFE]: There are no hardcoded credentials, sensitive file path references, or instructions that would lead to the exposure or exfiltration of user data.
- [SAFE]: The skill methodology (RIAEB) is used for psychological reframing and does not contain prompt injection patterns or attempts to override the underlying AI agent's safety guardrails.
- [SAFE]: A 'Boundary' section is explicitly defined, correctly advising that the skill is not a substitute for professional mental health treatment for conditions such as depression or clinical anxiety, which is a safety best practice for cognitive tools.
Audit Metadata