List conversations
List conversations with optional filtering, pagination, and ordering.
Authorizations
API key for authentication.
Query Parameters
Maximum number of conversations to return (1-250)
1 <= x <= 250Token for retrieving next page (from previous response)
Filter expression syntax.
Operators: =, !=, >, <, >=, <=, AND, OR
Values may be unquoted or double-quoted. Values containing spaces must be quoted.
Fields:
status- PENDING, IN_QUEUE, IN_PROGRESS, COMPLETED, FAILED, CANCELLED, DELETEDexternal_conversation_id- Your system's conversation IDcreate_time- ISO 8601 timestampoccurred_at- ISO 8601 timestampmetadata.{key}- Custom metadata fieldsmetric.{metric_id}- Filter by a metric's value (score filtering)
Metric-value filtering: a metric.{metric_id} predicate (e.g. metric.29Blkepvvx<"0.9";
>/</>=/<=/=/!= for float metrics, =/!= for string metrics) returns only
conversations whose metric matches, with every metric's value embedded inline (metric_values).
The metric type is inferred from the literal — a numeric literal is compared as a float metric.
When a metric predicate is present the request is served with keyset pagination, ordered
newest-first by creation time, and supports only AND alongside create_time (inclusive
bounds), agent_id, and metadata.{key}; other filter fields (including occurred_at) and any
non-default order_by are rejected with 400, and page_size is capped at 100.
Examples:
status=COMPLETEDcreate_time>"2025-11-01T00:00:00Z"status=COMPLETED AND occurred_at>="2025-11-01T00:00:00Z"external_conversation_id=external-call-abcmetric.29BlkepvvX19ebbLDB0y6Q<"0.5"
Sort field with optional - prefix for descending order.
Fields: create_time, occurred_at, status
Examples:
create_time(ascending)-create_time(descending, most recent first)-occurred_at(most recent conversations first)
Set to metric_breakdown to return an aggregate of one metric's scores
grouped by a customer_metadata key (e.g. vendor), computed over the whole
scored monitoring corpus, instead of the conversation list. Requires
metric_id and group_by_metadata; the response is a metric-breakdown object
({view, metric_id, group_by_metadata, aggregation, breakdown:[{metadata_value, value, count}], total_count}).
metric_breakdown Metric to aggregate when view=metric_breakdown, or the metric whose full
outputs should be embedded when include=metric_outputs.
Set to metric_outputs to embed full outputs for metric_id on every conversation
in the returned page, or metric_values to embed every metric's value inline on each
conversation (keyset-paginated; page_size capped at 100). Omitted by default to keep
list payloads small.
metric_outputs, metric_values Required when view=metric_breakdown: the customer_metadata key to group by (e.g. nlp_provider).
Aggregation for view=metric_breakdown. Defaults to success (a YES/NO
success rate) for binary/string metrics and avg (numeric mean) for float
metrics.
success, avg Optional ISO-8601 lower bound (occurred_at) for view=metric_breakdown.
Optional ISO-8601 upper bound (occurred_at) for view=metric_breakdown.
Response
List of conversations, or a metric breakdown when view=metric_breakdown