Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Chinese-language design guide for filtering an agent's available tools, with no executable payload, persistence, or hidden data flow found.
Install this only if you want a Chinese-language skill that may influence tool selection before requests. Review or tighten the keyword rules before relying on it for sensitive tools such as shell execution, messaging, subagents, or business-document actions.
The trigger signals use broad everyday keywords such as “文档”, “消息”, and “执行”, which can cause the skill to activate in contexts the user did not intend. In a tool-filtering skill, accidental activation can expose or prioritize powerful tools like exec, process, write, or messaging tools, increasing the chance of unsafe tool availability or misrouting of actions.
The system prompt describes intent detection with keyword-only rules but does not define disambiguation, confidence thresholds, or exclusion criteria. That makes the filtering logic easy to over-trigger, and in this specific skill it can lead to sensitive capabilities being loaded based on weak lexical matches rather than clear user intent.
The manifest description is written entirely in Chinese and provides no indication that other languages are supported or that the language choice is intentional for a region-specific audience. Under the policy, locale-specific language should be opt-in or clearly justified to avoid excluding users unexpectedly.
整个技能说明及示例均固定为中文触发词和中文交互模式，没有说明是否支持其他语言，也未提供用户可选语言或显式的区域性适用范围。若该技能面向通用用户环境，这种隐含的单语言约束可能构成语言/locale 策略问题。
No suspicious patterns detected.
