zhangxuefeng-perspective
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a role-play configuration using research-backed data to ground the AI's persona. No malicious instructions or unsafe tool usage patterns were found.
- [PROMPT_INJECTION]: The skill contains role-play constraints to ensure the AI stays in character, but these are standard for persona skills and do not attempt to bypass core system safety filters. It includes a clear 'exit' command for the user.
- [EXTERNAL_DOWNLOADS]: The files reference numerous reputable Chinese media and information platforms (36kr.com, jiemian.com, bilibili.com) for biographical grounding. These are used as informational sources and do not involve remote code execution.
- [DATA_EXFILTRATION]: There is no evidence of unauthorized data access or transmission. The skill suggests using web search tools for informational purposes, which is standard functionality.
Audit Metadata