Security audit
Security checks for vulnerabilities and agentic risk
This skill is a narrow patent-search guidance skill for R&D users and does not show hidden execution, persistence, credential handling, or destructive behavior.
Before installing, confirm that you want a Chinese-oriented R&D patent-search assistant and that any PatSnap/Zhihuiya MCP data it uses is available under your account permissions. Review generated reports for cited sources, search scope, time range, and legal disclaimers, especially for FTO, infringement, or validity questions.
The skill metadata and body are entirely in Chinese and frame outputs in Chinese-oriented terms, with no indication that language should follow user preference. This can cause the agent to override the user’s chosen language, reducing usability and potentially causing misunderstandings in technical or legal contexts where precise wording matters.
The trigger scenarios are broad, natural-language requests without clear activation constraints, so the skill may be invoked for loosely related prompts and take over conversations unexpectedly. In an agent setting, ambiguous activation boundaries can cause incorrect tool selection, scope creep, and unintended disclosure or misuse of enterprise search resources.
No suspicious patterns detected.
