reflect

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists exclusively of markdown instructions for the agent and contains no executable scripts or external dependencies.
  • [SAFE]: There are no patterns of data exfiltration, credential harvesting, or unauthorized command execution.
  • [SAFE]: All proposed changes to memory or skills require explicit user approval, maintaining human control over the agent's self-modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 02:13 PM
Security Audit — agent-trust-hub — reflect