Security audit
Security checks for vulnerabilities and agentic risk
This is a product-planning skill that asks structured questions and produces a design document; its main caveat is broad activation guidance, not unsafe code or data behavior.
Install this if you want a Chinese-first product strategy and design-document workflow before implementation. Be aware it may steer pre-coding conversations into structured questioning and design planning, so invoke it intentionally or tell the agent when you want direct coding help instead.
The description is written as Chinese-only operational guidance without stating that language should follow user preference, which can lead to forced locale switching. While not a direct security exploit, it is a quality and policy risk because it may reduce transparency, confuse users, and make the skill behave unexpectedly in multilingual environments.
The trigger guidance is broad enough to match many ordinary brainstorming, planning, and idea-clarification requests, which can cause the skill to activate outside its narrow intended context. That increases the chance of misrouting users into a rigid workflow, reducing user control and potentially interfering with other more appropriate skills or direct assistance.
The instruction to proactively suggest this skill before any coding is overly expansive and can bias the agent toward invoking it in nearly all pre-implementation conversations. In practice, this can override user intent, create unnecessary friction, and cause unintended skill takeovers even when the user wants coding help, debugging, or another specialized workflow.
No suspicious patterns detected.
