Security audit
Security checks for vulnerabilities and agentic risk
This is a Chinese-language business writing helper made of Markdown templates, with no code execution, persistence, credential access, or hidden data handling.
Install this if you want Chinese-language templates for business plans, marketing plans, and competitor analysis. Be aware that generic prompts like asking for a marketing plan may route into this skill, and business details you provide will be used in the generated output, but the artifact itself does not run code or access external data.
The manifest description lists trigger words inline, but one of them is phrased broadly enough that it could be used in ordinary discussion of business topics rather than as an explicit skill invocation. The file does not provide exclusion conditions or activation boundaries to clarify when these phrases should or should not trigger the skill.
The entire skill, including the title, trigger phrases, prompt template, and example dialogue, is written exclusively in Chinese and implies operation in that language only. There is no indication that users may choose another language or that the Chinese-only constraint is a documented, region-specific requirement.
全文标题、触发词、提示模板和示例均固定为中文，未说明用户可选择其他语言或按用户输入语言响应。根据 SQP-3，若技能强制特定语言且没有用户选择或明确合理的区域性约束，属于语言/locale 策略问题。
The trigger phrase includes a very common natural-language expression ('帮我分析竞品'), which increases the chance that the skill activates unintentionally during ordinary conversation. In an agent setting, accidental activation can cause the model to switch into this skill's prompt template unexpectedly, leading to misrouting, irrelevant outputs, or disclosure of user business context to the wrong workflow.
The skill content is entirely hardcoded in Chinese and implicitly requires Chinese output without checking the user's language preference or documenting a locale restriction. This can cause misrouting, user confusion, and policy/application-layer errors in multilingual environments, though it is not directly a code-execution or data-exfiltration issue.
The trigger phrases are generic business/assistant requests rather than narrowly scoped commands, so the skill can activate on ordinary user prompts that merely ask for a marketing plan. That increases unintended routing and prompt hijack surface, because users may be forced into this rigid template when they did not explicitly opt into the skill.
The manifest description and user-facing instructions are entirely in Chinese, which effectively forces a specific language for interaction. There is no opt-in, alternative language option, or justification that this skill is intended only for a Chinese-language audience.
No suspicious patterns detected.
