Purpose
Protect the boundary
Detect risk before a message or action is executed.
Flag risk and force approval when needed.
Detect risk before a message or action is executed.
Reviews outgoing content, selected actions, and sensitive phrases.
Returns a decision and the reason for it.
Always defer to policy, jurisdiction, and human review for sensitive cases.
It reviews drafts and actions for risky language before anything is sent.
It explains when approval is needed.
Use it whenever regulated, sensitive, or high-stakes messaging needs human review.