Security audit
Security checks for vulnerabilities and agentic risk
This is a Markdown-only writing workflow skill with vague marketing language but no hidden code, persistence, credential handling, or dangerous actions.
Install only if you want a broad writing-planning assistant. Treat its generated claims about performance or accuracy as unverified, and share only reference materials you are comfortable having the agent read.
The description says the skill is suitable for '多种工作场景' and emphasizes broad professional support, but it does not define specific activation phrases, scope boundaries, or exclusion conditions. In a markdown skill file, this kind of broad applicability language can make it unclear when the skill should activate versus when it should not.
The description states '支持中文交互', and the document is framed around Chinese-language interaction without offering the user a language choice or clarifying that Chinese is optional. This can conflict with language/locale policy expectations when a skill appears to impose a specific language by default.
The phrase '提供自动化处理、专业的能力支持' is highly generic and overlaps with common assistant behavior rather than uniquely identifying this writing skill. Because the markdown does not pair this with specific trigger examples or exclusions, the activation boundary remains ambiguous.
The API key section says the skill requires no additional API key except for explicitly marked external APIs, which implies a purely local or self-contained markdown instruction workflow. However, later documentation states that content generation is performed via an LLM API and that LLM API support is required, creating a documentation-level contradiction about external service dependence.
No suspicious patterns detected.
