Read one test conversation
GET
One test conversation: the version it ran, how it ended (endCode), its turns and cost. In a run built from a real conversation nothing reaches the real world: wouldHaveDone lists what the agent would have sent, booked or written.
Requires an API key with the workspace:read scope.
apiKeyheader
AuthorizationBearer <token>A secret API key from the dashboard (Developers → API keys): vl_live_…, or vl_test_… for test mode. Each operation says which scope the key needs.
id*stringThe run id.
The test conversation
application/json- response
run*curl "https://api.vlayers.ai/v1/conversation-runs/$RUN_ID" \ -H "Authorization: Bearer $VOICELAYER_API_KEY"Example Response
{ "run": { "id": "0192f5fa-e3f2-7d40-8f90-0123456789ab", "agentId": "0192f5c2-6f7a-7b9c-8d1e-2f3a4b5c6d78", "versionId": "0192f5d5-9cad-7ebf-9a4b-5c6d7e8f9012", "inputKind": "replay", "purpose": "suite", "suiteRunId": "0192f5fa-e3f2-7d40-8f90-0123456789ab", "caseId": "0192f5f9-d2e1-7c3f-9e8f-90123456789a", "status": "finished", "endCode": "contact_turns_exhausted", "attempt": 1, "turns": 2, "costUsd": 0.012, "createdAt": "2026-10-20T15:04:05.000Z", "finishedAt": "2026-10-20T15:04:05.000Z", "expiresAt": "2026-10-20T15:04:05.000Z", "wouldHaveDone": [ { "turn": 1, "kind": "catalog_action", "name": "refund_order" } ], "transcript": null }}