Skip to main content
After POST /v1/sessions/{session_id}/next, check action before deciding what to display. A warning does not automatically end the conversation.

Possible warnings

Fallbacks and stops

degraded_reasons may include semantic_service_unavailable (semantic analysis unavailable), unresolved_previous_turn (previous question not resolved), or missing_user_text (a reply was absent when expected). summary_only_context remains in the schema, but a summary-only add_context is now a normal operation and is not flagged as degraded. stop_reason has two values:
  • max_turns_reached: the session has reached max_turns. The turn that answered the last allowed question is recorded, then the session is stopped; later /next calls return the same stop without analyzing the turn.
  • no_question_available: no question remains after hard exclusions.
Objective achievement produces a warning, not a stop.
Example: action: "ask" with warnings: ["objective_achieved"] means the objective is met and a question is still available. Your application decides whether to ask it or stop.