AI Webhooks
AI debug webhook
A diagnostic feed for a call that is still in progress. Set debug_webhook_url on your agent and every step it takes is posted to that URL as it happens: speech recognized, model called, tool invoked, context switched, error hit. Use it to work out why a call went the way it did — which tool the agent reached for, what came back, where a turn went wrong — or to react while the call is still live, such as paging a supervisor.
Each request carries call_info plus one or more event properties, where the property name is the event. One moment can produce several: evaluating a data_map webhook sends webhook, input, output, error_keys, and match together. Handle the properties you recognize and ignore the rest, since the set grows over time.
Setting the URL is what enables the feed. debug_webhook_level only widens it: at 2 it also carries conversation_add, llm_request, and llm_response, which fire on every turn and every model call.
postWebhookaiDebugWebhook
Payload
Example payload
{
"call_info": {
"content_type": "text/json",
"content_disposition": "post_data",
"conversation_type": "voice"
},
"session_start": {
"model": "gpt-4.1-mini",
"tts_engine": "elevenlabs",
"tts_voice": "rachel",
"language": "en-US",
"mode": "oart",
"rate": 24000,
"audio_format": "pcm16",
"tool_model": "gpt-4.1-mini",
"local_vad": true
},
"session_end": {
"reason": "normal",
"ended_by": "assistant",
"duration_ms": 37651,
"input_tokens": 5627,
"output_tokens": 119
},
"startup_hook": {
"duration_ms": 214,
"success": true,
"error": "execution failed"
},
"hangup_hook": {
"duration_ms": 186,
"success": true,
"error": "execution failed"
},
"speech_detect": {
"text": "I need a ride to the airport.",
"source": "final"
},
"barge": {
"barge_type": "normal",
"barge_elapsed_ms": 1240,
"interrupt_count": 2,
"audio_played_ms": 1240,
"barge_count": 2
},
"ai_completion": {
"type": "normal"
},
"filler": {
"text": "Let me check on that.",
"filler_type": "function"
},
"attention_timeout": {
"timeout_ms": 15000
},
"hold": {
"event": "hold",
"timeout": 30
},
"function_call": {
"function": "get_weather",
"duration_ms": 412,
"shadow": true,
"phase": "post_prompt"
},
"function_loop": {
"function": "get_weather",
"type": "repeated",
"count": 3
},
"swaig_call": {
"command_name": "get_weather",
"command_arg": "{\"city\":\"San Francisco\"}",
"epoch_time": 1694541334,
"native": true,
"url": "https://example.com/tools/get_weather",
"mcp_error": true
},
"tool_distill": {
"function": "get_weather",
"raw_chars": 8412,
"distilled_chars": 612,
"model": "gpt-4.1-mini"
},
"error": {
"error": "ai: webhook error",
"errorText": "Webhook obj must have output or expressions"
},
"webhook_log": {
"request_url": "https://example.com/weather?city=Springfield"
},
"webhook_fail": {
"call_info": {
"content_type": "text/json",
"content_disposition": "post_data",
"conversation_type": "voice"
},
"error": "webhook error",
"request_url": "https://example.com/tools/get_weather",
"attempts": 3,
"parse_error": true,
"protocol_error": true,
"curl_code": 28,
"http_code": 502
},
"step_change": {
"from_step": "greet",
"to_step": "collect_address",
"to_index": 1,
"trigger": "ai_function"
},
"context_change": {
"to_context": "billing",
"from_context": "default",
"trigger": "ai_function"
},
"gather_start": {
"context": "default",
"step": "collect_address",
"step_index": 2,
"output_key": "address",
"total_questions": 3
},
"gather_question": {
"context": "default",
"step": "collect_address",
"step_index": 2,
"key": "street",
"question_type": "string",
"requires_confirm": true
},
"gather_answer": {
"context": "default",
"step": "collect_address",
"step_index": 2,
"key": "street",
"confirmed": true
},
"gather_reject": {
"context": "default",
"step": "collect_address",
"step_index": 2,
"key": "street",
"reason": "missing_answer"
},
"gather_complete": {
"context": "default",
"step": "collect_address",
"step_index": 2,
"output_key": "address",
"answered": 3,
"completion_action": "next_step"
},
"conversation_add": {
"role": "assistant",
"content": "Your ride is booked for 6pm.",
"lang": "en-US",
"tokens": 53,
"timestamp": 1694541297950440
},
"llm_request": {
"model": "gpt-4.1-mini",
"prompt_tokens": 412,
"conversation_tokens": 5215
},
"llm_response": {
"model": "gpt-4.1-mini",
"output_tokens": 53,
"duration_ms": 836,
"success": true
},
"llm_error": {
"event": "llm_retry",
"model": "gpt-4.1-mini",
"token_count": 128000,
"limit": 120000,
"errors": 4,
"reasoning_effort": "high"
},
"consolidation": {
"event": "consolidation",
"token_count": 121400,
"limit": 120000
},
"summarize_start": {
"mode": "function",
"model": "gpt-4.1-mini"
},
"post_prompt": {
"url": "https://example.com/reports"
},
"text_normalize": {
"direction": "itn",
"language": "en",
"original": "twenty twenty six",
"normalized": "2026"
},
"voice_error": {
"event": "voice_config_error",
"engine": "elevenlabs",
"voice": "rachel",
"fallback_engine": "gcloud",
"fallback_voice": "en-US-Neural2-J"
},
"warning": {
"event": "excessive_prompt_tokens",
"tokens": 82000
},
"oart_user_transcript": {
"text": "I need a ride to the airport."
},
"oart_assistant_transcript": {
"text": "Your ride is booked for 6pm."
},
"oart_response_done": {
"latency": 836,
"utterance_latency": 934,
"audio_latency": 1106,
"acoustic_latency": 1180,
"status": "completed"
},
"inner_dialog": {
"text": "The caller has not given a pickup address yet."
}
}Response
Webhook received