Security audit
Security checks for vulnerabilities and agentic risk
This skill is a Chinese WeChat article drafting and formatting guide with no executable code or hidden high-impact behavior.
Install this if you want help drafting WeChat public-account articles in its custom Markdown module format. Review generated public-facing content, links, public-account IDs, and external image URLs before publishing.
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.
# 微信公众号排版模块完整参考
本文档列出全部 40 个 fenced block 模块的精确字段说明和写法示例。
The skill description is broadly scoped to 'when the user needs to write, edit, or format WeChat public account articles' without clear exclusion boundaries. This can cause the agent to invoke the skill for loosely related requests and let the skill steer outputs in contexts where a narrower or safer capability should apply, increasing prompt-routing errors and unintended file generation.
The tags include broad natural-language trigger phrases like '写一篇公众号' and generic content categories, which may overlap with ordinary user speech. In agentic systems that use tags for retrieval or invocation, this raises the chance of accidental activation and inappropriate delegation, especially for general writing requests that only partially match the skill.
The title explicitly frames the skill reference around WeChat official-account formatting in Chinese, which establishes a fixed language/locale context. Across the document, the instructions and examples assume Chinese-language output and a China-specific publishing platform without stating that this is an intentional locale-specific tool or offering alternatives.
No suspicious patterns detected.