chinese-doc-style
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill does not contain any executable code, scripts, or commands. It consists entirely of natural language instructions and configuration for an AI agent.
- [EXTERNAL_DOWNLOADS]: The skill references external documentation guidelines from well-known services, specifically a copywriting guide on GitHub and a style manual on Wikipedia. These are static informational links used for reference only and do not involve remote code execution or package installation.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and modify user-supplied text. It includes specific guardrails to protect technical content (such as shell commands, URLs, paths, and code blocks) from being modified. It also explicitly instructs the agent to preserve author stance and avoid inventing data. The risk of processing untrusted data is mitigated by these instructions and the limited scope of documentation formatting. Evidence: 1. Ingestion point: User-provided documents (SKILL.md). 2. Boundary markers: Instructions to protect technical content and preserve author stance. 3. Capability: The skill mentions the ability to edit files. 4. Sanitization: Instructions to keep identifiers and code blocks unchanged.
Audit Metadata