Get Conversations

{ "conversations": [ { "agent_id": "<string>", "conversation_id": "<string>", "start_time_unix_secs": 123, "call_duration_secs": 123, "message_count": 123, "status": "initiated", "call_successful": "success", "branch_id": "<string>", "version_id": "<string>", "agent_name": "<string>", "termination_reason": "", "call_success_score": 123, "transcript_summary": "<string>", "call_summary_title": "<string>", "main_language": "<string>", "conversation_initiation_source": "unknown", "tool_names": [ "<string>" ], "direction": "inbound", "rating": 123, "sentiment_analysis": { "overall_label": "positive", "overall_sentiment_score": 0, "overall_frustration_score": 0.5, "min_user_sentiment_score": 0, "max_user_frustration_score": 0.5, "num_scored_user_turns": 1 }, "data_collection_results": {}, "evaluation_criteria_results": {}, "tag_ids": [ "<string>" ] } ], "has_more": true, "next_cursor": "<string>" }

Headers

x-api-key

string | null

Your API key for programmatic access. Send it in the x-api-key header; you can create and manage keys from the Developers page.

Query Parameters

cursor

string | null

Used for fetching next page. Cursor is returned in the response.

agent_id

string | null

Agent id (agent_…) or speech engine external id (seng_), resolved to the same underlying resource.

Example:

"agent_3701k3ttaq12ewp8b7qv5rfyszkz"

visited_agent_ids

string[] | null

Filter conversations where any of these agents participated. Can not exceed 50 values.

Maximum array length: 50

visited_agent_branch_ids

string[] | null

Filter conversations where any of these agent branches participated. Can not exceed 50 values.

Maximum array length: 50

triggered_procedure_ids

string[] | null

Filter conversations where any of these procedures were triggered. Can not exceed 50 values.

Maximum array length: 50

call_successful

enum<string> | null

The result of the success evaluation

Available options:

success,

failure,

unknown,

error

Example:

"success"

call_start_before_unix

integer | null

Unix timestamp (in seconds) to filter conversations up to this start date.

call_start_after_unix

integer | null

Unix timestamp (in seconds) to filter conversations after to this start date.

call_duration_min_secs

integer | null

Minimum call duration in seconds.

call_duration_max_secs

integer | null

Maximum call duration in seconds.

rating_max

integer | null

Maximum overall rating (1-5).

Required range: 1 <= x <= 5

rating_min

integer | null

Minimum overall rating (1-5).

Required range: 1 <= x <= 5

has_feedback_comment

boolean | null

Filter conversations with user feedback comments.

user_id

string | null

Filter conversations by the user ID who initiated them.

evaluation_params

string[] | null

Evaluation filters. Repeat param. Format: criteria_id:result. Example: eval=value_framing:success

data_collection_params

string[] | null

Data collection filters. Repeat param. Format: id:op:value where op is one of eq|neq|gt|gte|lt|lte|in|exists|missing. For in, pipe-delimit values.

data_collection_ids

string[] | null

Data collection field IDs to include in each conversation summary. Repeat param. When omitted, data_collection_results is not returned.

evaluation_criteria_ids

string[] | null

Evaluation criteria IDs to include in each conversation summary. Repeat param. When omitted, evaluation_criteria_results is not returned.

tool_names

string[] | null

Filter conversations by tool names used during the call.

tool_names_successful

string[] | null

Filter conversations by tool names that had successful calls.

tool_names_errored

string[] | null

Filter conversations by tool names that had errored calls.

include_invalid_tool_calls

boolean

default:false

Also match tool calls that never ran.

main_languages

string[] | null

Filter conversations by detected main language (language code).

page_size

integer

default:30

How many conversations to return at maximum. Can not exceed 100, defaults to 30.

Required range: 1 <= x <= 100

summary_mode

enum<string>

default:exclude

Whether to include transcript summaries in the response.

Available options:

exclude,

include

conversation_initiation_source

enum<string> | null

default:unknown

Enum representing the possible sources for conversation initiation.

Available options:

unknown,

android_sdk,

node_js_sdk,

react_native_sdk,

react_sdk,

js_sdk,

python_sdk,

widget,

sip_trunk,

twilio,

exotel,

genesys,

avaya,

audiocodes,

swift_sdk,

whatsapp,

twilio_sms,

flutter_sdk,

zendesk_integration,

slack_integration,

telegram_integration,

intercom_integration,

freshdesk_integration,

salesforce_integration,

template_preview,

genesys_bot_connector,

subagent_tool

text_only

boolean | null

conversation_product_type

enum<string> | null

Restrict results to a single conversation product surface. Which product surface owns this agent document.

Available options:

agents,

speech_engine

branch_id

string | null

Filter conversations by branch ID.

version_id

string | null

Filter conversations by version ID.

parent_conversation_id

string | null

Filter conversations by parent conversation ID for subagent conversations.

topic_ids

string[] | null

Filter conversations by topic IDs assigned during topic discovery.

exclude_statuses

enum<string>[] | null

Exclude conversations with the given statuses. Useful for hiding in-progress / processing conversations from list views.

Available options:

initiated,

in-progress,

processing,

done,

failed

tag_ids

string[] | null

Filter conversations by conversation tag IDs assigned via the conversation-tags endpoints.

workflow_node_entered_id

string | null

Filter conversations to only those that entered the given node.

termination_reasons

string[] | null

Filter conversations by their stored termination_reason (metadata.termination_reason). Repeat param to match any of several.

guardrail_types

enum<string>[] | null

Filter to conversations where a guardrail of any of these types triggered (metadata.triggered_guardrails.guardrail_type). Repeat param to match any of several.

Available options:

custom,

prompt_injection,

self_harm_intent,

violence_graphic,

sexual,

violence,

harassment,

sexual_minors,

self_harm,

self_harm_instructions,

harassment_threatening,

hate,

hate_threatening,

profanity,

religion_or_politics,

medical_and_legal,

guardrail

custom_guardrail_names

string[] | null

Filter to conversations where a custom guardrail with any of these names triggered (metadata.triggered_guardrails.guardrail_name). Only custom guardrails carry a name. Repeat param to match any of several.

sort_direction

enum<string>

default:desc

The direction to sort conversations by call start time. Defaults to descending (newest first).

Available options:

asc,

desc

Response

Successful Response

conversations

ConversationSummaryResponseModel · object[]

required

Show child attributes

has_more

boolean

required

next_cursor

string | null

Agents Platform

API Reference

Agents Insights

Agents Workspace Analytics

Outbound Calls

Conversational AI

Voices

Phone Numbers

workspace

access:all

Time Zones

Headers

Query Parameters

Response