Security audit
Security checks for vulnerabilities and agentic risk
This skill is a text-analysis and short-comment generator for Chinese posts, with no code execution, data access, persistence, or hidden external behavior found.
Before installing, understand that this is a specialized Chinese-text workflow. It may activate on broad requests like article analysis or support/opposition comments, so users should invoke it when they specifically want AI-style writing analysis or short grounded comments for Chinese posts.
The trigger examples include very broad phrases such as ‘分析这篇文章’ and ‘帮我从正反两面评论’, which overlap with common user intents unrelated to this narrow skill. In an agentic system, this can cause unintended invocation, routing ordinary analysis/commentary requests into a specialized workflow that performs extra classification and persuasion-oriented output the user did not explicitly request.
The description states that the skill analyzes 普通中文博文, and later examples and supported content are all constrained to Chinese text. This is a language/locale restriction presented as a fixed requirement, but the file does not explicitly frame it as an opt-in user choice or justify the constraint as a region-specific requirement.
No suspicious patterns detected.
