personality

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a persona-switching system with 26 predefined identities, allowing users to toggle between different communicative styles via the /personality command.
  • [SAFE]: Persistent state management is handled by writing to a local SOUL.md file, using specific comment blocks to prevent overwriting unrelated content.
  • [SAFE]: The instructions explicitly mandate that personality changes must never reduce correctness, safety, or policy compliance, and that technical data (code, paths, error details) must remain unstyled and literal.
  • [SAFE]: Personas mimicking sensitive professional roles, such as medical or legal consultants, include mandatory disclaimers clarifying that the output is informational only and not professional advice or diagnosis.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 09:02 PM
Security Audit — agent-trust-hub — personality