Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only Chinese analysis helper with broad trigger phrases, but it does not request tool access, persistence, credentials, or code execution.
Installers should understand that this skill is intended for Chinese-language structured analysis. Use it for user-provided data, decisions, or problem descriptions, and consider asking for confirmation if a generic phrase like '帮我决策' appears in ambiguous conversation.
The phrase '诊断问题' is ambiguous and applicable to many everyday scenarios, which makes accidental or over-broad skill invocation likely. In an agent environment, ambiguous triggers can cause unintended handling of sensitive, unrelated, or higher-risk user requests under this skill's instructions.
The phrase '诊断问题' is ambiguous and applicable to many everyday scenarios, which makes accidental or over-broad skill invocation likely. In an agent environment, ambiguous triggers can cause unintended handling of sensitive, unrelated, or higher-risk user requests under this skill's instructions.
The skill content is written to produce Chinese-language output without giving the user a language choice, which can override user expectations or platform language settings. While not a direct code-execution risk, it can degrade trust, cause misunderstanding of analytical results, and create unsafe outcomes if users misinterpret important recommendations due to language mismatch.
The trigger phrases are broad enough to match common user requests, which can cause this skill to activate unexpectedly in ordinary conversations. That increases the chance of prompt hijacking or misrouting user intent, especially because '分析数据' is a generic request many users may naturally make outside a narrowly scoped skill invocation flow.
The trigger phrases "帮我决策" and "帮我做决定" are common, everyday-language requests that are likely to appear in normal conversation, making accidental or overly broad activation plausible. In a decision-support skill, unintended activation can cause the agent to switch into a structured advisory mode without clear user intent, which may lead to inappropriate guidance or context confusion.
The trigger phrases 诊断问题 and especially 帮我分析问题 are broad, natural-language phrases likely to appear in ordinary conversation, which can cause the skill to activate unintentionally. In an agent setting, over-broad activation increases the risk of prompt hijacking, misrouting user requests, or exposing internal skill behavior when the user did not explicitly intend to invoke this diagnostic workflow.
The natural-language instructions and trigger phrases are all Chinese, and the file does not indicate that users may choose another language or that the skill is intentionally limited to a Chinese-language audience. This can amount to a language policy issue when a skill implicitly forces a locale without user opt-in.
No suspicious patterns detected.
