wujun-perspective

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The wujun-perspective skill consists entirely of text-based instructions for persona adoption and reasoning workflows. It contains no scripts, binaries, or executable commands that interact with the host system.
  • [SAFE]: Analysis of the skill body and referenced files revealed no attempts at data exfiltration, credential harvesting, or access to sensitive local file paths.
  • [SAFE]: The skill does not contain prompt injection patterns or instructions to bypass safety guidelines. Its structure is designed to guide the AI agent's analysis style without overriding underlying safety protocols.
  • [SAFE]: References to external sources (such as GitHub repositories) are descriptive and informative, pointing to the project's own documentation and research materials.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 02:53 AM
Security Audit — agent-trust-hub — wujun-perspective