memory-triage
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates a mandatory 'Gate 4
- SAFE' check that requires the agent to identify and exclude sensitive information, such as authentication tokens or credentials, from being stored in long-term memory.
- [SAFE]: No unauthorized network requests, external downloads, or remote code execution patterns were identified. All tool usage is restricted to memory management functions.
- [SAFE]: The instructions provide clear, safety-oriented guidelines for data extraction, emphasizing entity grouping and the avoidance of unverified inferences or assistant-attributed characterizations.
Audit Metadata