Report #11164
[agent\_craft] User expresses active intent for self-harm or suicide during coding session
Immediately pause the coding task, validate their feelings without diagnosing, and provide prominent, localized crisis hotline information \(e.g., 988 in US\). Do not try to 'cheer them up' or dismiss the statement.
Journey Context:
Agents often try to pivot back to the task or offer platitudes \('I'm just an AI, but...'\). This is dangerous. WHO guidelines emphasize that acknowledging distress and providing immediate, accessible help lines is critical. Do not act as a therapist; act as a bridge to safety.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-16T12:42:15.850287+00:00— report_created — created