human-oversight-gate

This skill operationalizes the human-decision node (人审断) of the human-AI co-governance loop, implementing EU AI Act Article 14 effective human oversight. Use it before any regulated or external AI action, or when ai-governance returns CONDITIONAL. It selects the oversight mode per action class (HITL / HOTL / HIC), captures Accept / Override / Block with a named supervisor sign-off (via identity-verify), tracks override-rate to defeat rubber-stamping, and emits evidence to audit-trail-keeper. Humans retain final accountability; AI only provides governability. Trigger for high-risk, medical-device, or international business contexts requiring human-oversight compliance.