agent-soul

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and data-driven, using local markdown files to define personality archetypes. It does not execute any code, scripts, or network operations.
  • [SAFE]: It explicitly defines 'Protected surfaces' (plans, diffs, tool calls, and security findings) that must remain neutral, preventing the persona from interfering with the agent's core functionality or safety protocols.
  • [SAFE]: A 'Serious mode' is implemented to allow users to instantly mute the persona and revert to baseline behavior, ensuring user control over the agent's communication style.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 10:43 AM