zhangjuzheng-perspective
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate persona extension for historical administrative and management guidance.- [PROMPT_INJECTION]: The roleplay guidelines and Agentic Protocol are used solely to establish the Zhang Juzheng persona. No instructions to bypass safety guidelines, disregard previous rules, or extract system prompts were found.- [DATA_EXFILTRATION]: The skill does not access sensitive file paths (such as SSH keys or environment variables) and does not perform network operations to exfiltrate data.- [REMOTE_CODE_EXECUTION]: No remote code execution patterns, automated package installations, or piped shell commands were detected. The installation instruction in the documentation is standard and requires user action.- [OBFUSCATION]: All instructions, examples, and research data are provided in clear text. There are no encoded strings, zero-width characters, or hidden URLs designed to evade detection.
Audit Metadata