critical-thinking-mode
Pass
Audited by Gen Agent Trust Hub on Mar 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructional frameworks for reasoning and questioning. It does not contain any malicious instructions, prompt injection attempts, or hidden code.
- [COMMAND_EXECUTION]: The skill instructions explicitly state that the agent should 'not propose solutions' and is 'not here to make code edits.' This restriction limits the agent's capability to execute potentially dangerous commands or modify the user's environment.
- [EXTERNAL_DOWNLOADS]: While discovery tools like
web/fetchare listed in the reference metadata for fact-gathering, the skill does not hardcode any remote URLs, third-party dependencies, or payloads.
Audit Metadata