Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only publishing data-analysis skill with disclosed analytics behavior and no bundled code or hidden execution path.
Review whether you want a skill that may be invoked for general data-report requests; otherwise its described behavior is proportionate for publishing analytics. If connected to real platform accounts later, ensure collection is user-authorized, scoped, and does not store personal reader data.
The trigger description is broad enough to match common tasks like generating data reports or analyzing feedback, which can cause the skill to activate outside its intended scope. In a multi-agent system, over-broad routing can expose unrelated user data or cause this agent to collect/process platform data when a narrower, safer skill should have handled the request.
L003 的自然语言描述整体以中文固定表述技能行为，且文档未说明是否可根据用户偏好切换语言或输出本地化内容。按规则，若技能隐含强制特定语言/locale 而无用户 opt-in 或明确合理依据，属于自然语言政策问题。
No suspicious patterns detected.
