learn-anything
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines instructional patterns for an educational tutor agent. It focuses on pedagogical techniques such as active retrieval, Socratic questioning, and scaffolding to assist learners.
- [DATA_EXFILTRATION]: While the skill instructs the agent to read user-provided sources, files, or code to verify claims, there are no network operations or external URLs that would suggest data exfiltration or unauthorized sharing of user information.
- [PROMPT_INJECTION]: The instructions include a provision to exit coaching and answer normally if the user requests it. This is a functional feature for user control and does not attempt to bypass core safety guardrails or ethics protocols.
Audit Metadata