Get transcript
Get a session's conversation transcript (id from list_calls, or a run's sessionId from get_test_results). Defaults to a compact, ready-to-read text rendering of the call (turns merged, each tool call shown as one line with a clipped result). Pass format:"json" for structured message rows, as JSON inside an untrusted-data block (tool payloads clipped unless full:true; each row's metadata, such as interruption times and word timings, with include_metadata:true). Returns session_not_found if the id doesn't resolve to a session you can access. For everything recorded about a call (timing, spans, recordings, metrics, carrier events), use get_call_bundle.
Authorization
bearerAuth A personal access token from Settings > API tokens.
In: header
Header Parameters
A key you choose, such as a UUID, that makes this POST safe to retry. Within 24 hours the same key and request answers with the first successful answer and runs nothing, the same key with a different request answers 422 idempotency_key_reused, and a key whose request is still running answers 409 idempotency_key_in_use. A failed request keeps nothing, so the same key runs it again.
1 <= length <= 255Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
application/problem+json
application/problem+json
application/problem+json
application/problem+json
application/problem+json
application/problem+json
application/problem+json
curl -X POST "https://example.com/operations/get_transcript" \ -H "Content-Type: application/json" \ -d '{ "session_id": "string" }'{}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}{
"type": "http://example.com",
"title": "string",
"status": 0,
"detail": "string",
"code": "string",
"errors": [
{
"path": "string",
"message": "string"
}
],
"estimate": {},
"retryAfterSeconds": 0
}Runs overview POST
Get the aggregate eval-run trend for the agent you're editing over the last N days (default 7): overall success rate, number of runs, average run duration, and the top failed scenarios (by failure count). Use this to answer "how is the agent doing across all its tests?" or "which scenarios are failing most?", it summarises every finished run in the window, not a single run. Each run counts by its verdict, as on Test analytics: an operator override wins, a passed run whose critical metric failed counts as a failure, and errored, cancelled, unfinished and still-scoring runs are left out of the run count, the success rate and the failure counts. Workspace-wide scenarios count when they ran against this agent. Returns an empty summary when the agent has no finished runs in the window. Differs from `get_test_results`: this is the trend ACROSS many runs over a window, use get_test_results instead for the per-test detail (verdict + failed-rubric reasoning) of one specific run or batch.
Get workspace settings POST
Read a workspace's settings: name, URL slug, timezone, language, how many tests run at once in a batch (evalConcurrency), and the privacy and retention settings (call retention, audit log retention, what transcripts keep, which personal details are hidden, call recording, who may download recordings, what reaches support tools and analytics, IP capture, session replay, and zero data retention). Read this before changing any of them.