skill-zhihu-publisher
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and metadata do not contain any malicious patterns. The functionality is focused on browser automation for a specific platform (Zhihu) using a shared script.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies user-provided content (titles and descriptions) as inputs for the publishing process. While this creates a standard entry point for data processing, the instructions handle this through localized execution variables and do not exhibit exploitable patterns.
Audit Metadata