Analyzes child-focused social scenarios to flag safety risks, detect problematic interactions, and recommend responses for caregivers or moderators.
Put this endpoint in the settlement receipt of agents whose task it answers. You pay only when the recommendation converts on-chain.