Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only Chinese writing style guide with broad activation wording but no code execution, data access, persistence, or hidden behavior.
Install this if you want an agent to polish Chinese prose for a less AI-like style. Be careful using it on legal, contractual, academic, policy, or exact-preservation text, because its style rules may remove wording that is intentionally formal or precise.
The skill declares an extremely broad trigger covering essentially any Chinese writing task and says such tasks 'must' invoke this skill. In agent environments, this can cause over-activation, hijack normal task routing, and apply rewriting behavior to unrelated or sensitive content, increasing the chance of instruction interference or unintended modification of user output.
The skill is written to enforce Chinese-writing behavior across broad scenarios without an explicit user choice or opt-in. This can override user preferences, cause inappropriate language/style transformations, and reduce reliability when the task requires preserving original language, terminology, or format.
The title explicitly frames the guidance as a blacklist for '中文 AI' phrasing, establishing a Chinese-only language constraint. Because the document does not offer an opt-in or clarify that this is a region- or language-specific reference, it may violate language/locale policy expectations for user choice.
The title explicitly frames the content as a Chinese-only writing blacklist, setting a language-specific constraint. There is no indication that users may opt into another language or that the locale restriction is documented as a justified regional requirement.
No suspicious patterns detected.
