Risk Detection API

Listen → Detect → Flag → Alert

Monitor transcripts and clinical text for suicidal ideation, self-harm, and escalation — surfacing timestamped, evidence-backed risk flags in real time.