Tests And Evaluations
Get Agent Evaluation Conversation
Get an agent evaluation conversation to inspect one test attempt’s transcript, judgement, extracted fields, and available recording or latency data.
GET
Get Agent Evaluation Conversation
Read only
Get an agent evaluation conversation to inspect one test attempt’s transcript, judgement, extracted fields, and available recording or latency data.
Get Agent Evaluation Conversation
Get the conversation ID from Get Agent Evaluation Run. Use the same agent and run IDs.Minimal request
No request body is needed. Supply all three IDs in the path.Evaluation conversation limits
Text runs do not exercise speech recognition or synthesis. An absent recording or audio measurement does not establish an audio-quality result. Useerror_message to investigate an attempt that could not complete; distinguish the error outcome from a judged fail.
The original test name and type fall back to Deleted test and simulation only if the test record no longer exists. Review the transcript and result fields rather than inferring the original test from those fallback labels.
Verify the result
A200 response contains the conversation in data. Read status, outcome, summary, and failure_reason alongside transcript, criteria, params, outcomes, and builtin. Unfinished conversations can have no outcome and empty judgement arrays.
Errors and recovery
404 means the conversation does not belong to the supplied run and agent, or is not available in your workspace. Correct the IDs before retrying. Read requests are safe to repeat.
Missing, expired, or invalid API credentials return 401. See API authentication.
Related endpoints
- List Agent Evaluation Tests
- Create Agent Evaluation Test
- Start Agent Evaluation Run
- Get Agent Evaluation Run
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
Agent ID from List Agents.
Run ID from Start or List Agent Evaluation Runs.
Conversation ID from Get Agent Evaluation Run.