Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Zhihu-focused Chinese writing guide with no executable code, hidden access, persistence, or credential handling.
Install this if you want an agent to draft Chinese long-form content in a Zhihu style. Be aware it may bias output toward Zhihu-specific engagement and formatting patterns even for other Chinese platforms; review citations, claims, and platform rules before publishing.
The README describes the skill as being for writing Zhihu answers and articles, and later extends this to long-form Chinese social media content. This indicates a specific language/locale orientation, but the file does not explicitly offer opt-in language choice or explain the locale restriction as a documented regional limitation.
The activation condition is broad enough to trigger on 'any long-form Chinese social media content,' which exceeds the stated Zhihu-specific purpose and can cause the skill to be applied in unrelated contexts without clear user intent. In agent systems, overbroad activation can lead to inappropriate capability invocation, undesired behavioral shaping, and accidental platform-specific optimization being applied where it does not belong.
The skill is framed as a general writing engine but its instructions and examples throughout the file require Zhihu-specific Chinese writing conventions, Chinese phrases, and platform-native structure. There is no explicit user choice or opt-in for language/locale behavior, which can violate language policy requirements when a skill constrains output to a specific language by default.
This markdown file contains only Chinese-language instructions and output content, and there is no indication that users can opt into another language or that the skill is explicitly limited to a Chinese-only context. Under the policy rule, forcing a specific language without user choice is a natural-language policy concern.
No suspicious patterns detected.
