Security audit
Security checks for vulnerabilities and agentic risk
This is a simple HR handbook drafting skill with clear limits and no evidence of hidden execution, data access, persistence, or exfiltration.
Before installing, treat this as a drafting assistant for HR handbook language, not legal advice. Provide only the company context needed for the draft, avoid unnecessary personal employee details, and have final policies reviewed by a qualified HR or labor-law professional.
The trigger keywords are broad HR phrases such as '员工手册', '考勤制度', and '离职流程' that can appear in ordinary conversation without a clear request to invoke this skill. This can cause over-activation, routing unrelated HR/legal discussions into a skill that generates policy text, increasing the chance of inappropriate or low-context output being treated as authoritative.
The natural-language examples are open-ended and do not define clear boundaries for when the skill should activate versus when the system should ask clarifying questions or avoid activation. In a compliance-adjacent HR context, this broad matching can misroute general labor-law or employee-relations questions into a document-drafting workflow, creating risk that users rely on generated policy language without sufficient context.
No suspicious patterns detected.
