Skip to main content
GET
List runs

Authorizations

X-API-Key
string
header
required

API key for authentication

Query Parameters

filter
string

Filter expression syntax.

Supported fields: status, agent_id, persona_id, test_set_id, create_time, update_time, created_by, tag, metric.{metric_id}

Operators: =, !=, >, <, >=, <=, AND, OR

Values may be unquoted or double-quoted. Values containing spaces must be quoted (e.g., status="IN PROGRESS").

Metric-value filtering: a metric.{metric_id} predicate (e.g. metric.29Blkepvvx<"0.9"; >/</>=/<=/=/!= for number metrics, =/!= for string metrics; type inferred from the literal) returns only runs whose metric matches, with each run's per-metric averages embedded inline (metric_averages). On this path the request is keyset-paginated, ordered newest-first by creation time, and supports only AND alongside create_time (inclusive bounds), status, agent_id, test_set_id, and persona_id; other fields and any non-default order_by are rejected with 400, and page_size is capped at 100.

page_size
integer
default:50

Maximum number of results per page

Required range: 1 <= x <= 1000
page_token
string

Opaque pagination token from previous response

order_by
string
default:-create_time

Sort order specification.

Format: field or -field (descending)

Supported fields: create_time, update_time, status, agent_id, persona_id, test_set_id

include
enum<string>

Set to metric_averages to embed each run's average value per metric inline. Served by the keyset engine (newest-first, page_size capped at 100).

Available options:
metric_averages

Response

List of runs

runs
object[]
required
next_page_token
string

Token for fetching the next page of results

Example:

"eyJvZmZzZXQiOiA1MH0="