curl --request POST \
--url https://app.egma.ai/v1/runs \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"suiteId": "ste_01M0E4J0BBE1FVDVTZ1BSS5C97",
"agentId": "agt_01M0E4J0BBE1FVDVTZ1BSS5C97",
"connectionId": "con_01M0E4J0BBE1FVDVTZ1BSS5C97"
}
'{
"id": "<string>",
"projectId": "<string>",
"suiteId": "<string>",
"suiteName": "<string>",
"suiteDeleted": true,
"name": "<string>",
"status": "pending",
"agentId": "<string>",
"connectionId": "<string>",
"connectionName": "<string>",
"agentPlatform": "<string>",
"connectionType": "<string>",
"accessVariant": "<string>",
"modality": "voice",
"productLabel": "<string>",
"environment": "<string>",
"agentVersion": 123,
"expectedSimulationCount": 123,
"completedCount": 123,
"failedCount": 123,
"canceledCount": 123,
"simulationCounts": {
"queued": 123,
"claimed": 123,
"running": 123,
"completed": 123,
"failed": 123,
"canceled": 123
},
"finishedCount": 123,
"gradableCount": 123,
"gradedCount": 123,
"resultsUrl": "<string>",
"createdAt": "2023-11-07T05:31:56Z",
"startedAt": "2023-11-07T05:31:56Z",
"finishedAt": "2023-11-07T05:31:56Z"
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}Create a run
Run every active test in a suite against one agent connection. Execution and grading continue after this request returns. Keep the returned id to follow progress and inspect results.
curl --request POST \
--url https://app.egma.ai/v1/runs \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"suiteId": "ste_01M0E4J0BBE1FVDVTZ1BSS5C97",
"agentId": "agt_01M0E4J0BBE1FVDVTZ1BSS5C97",
"connectionId": "con_01M0E4J0BBE1FVDVTZ1BSS5C97"
}
'{
"id": "<string>",
"projectId": "<string>",
"suiteId": "<string>",
"suiteName": "<string>",
"suiteDeleted": true,
"name": "<string>",
"status": "pending",
"agentId": "<string>",
"connectionId": "<string>",
"connectionName": "<string>",
"agentPlatform": "<string>",
"connectionType": "<string>",
"accessVariant": "<string>",
"modality": "voice",
"productLabel": "<string>",
"environment": "<string>",
"agentVersion": 123,
"expectedSimulationCount": 123,
"completedCount": 123,
"failedCount": 123,
"canceledCount": 123,
"simulationCounts": {
"queued": 123,
"claimed": 123,
"running": 123,
"completed": 123,
"failed": 123,
"canceled": 123
},
"finishedCount": 123,
"gradableCount": 123,
"gradedCount": 123,
"resultsUrl": "<string>",
"createdAt": "2023-11-07T05:31:56Z",
"startedAt": "2023-11-07T05:31:56Z",
"finishedAt": "2023-11-07T05:31:56Z"
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}Authorizations
An Egma API key.
Query Parameters
Project to act in. A project-scoped API key already identifies its project.
1Body
The active, non-empty test suite to execute in full. It must belong to the selected project.
1"ste_01M0E4J0BBE1FVDVTZ1BSS5C97"
The Egma agent to test, not its Retell provider ID or LiveKit dispatch name.
1"agt_01M0E4J0BBE1FVDVTZ1BSS5C97"
An active connection on that agent in the same project. Its modality determines whether simulations use voice or chat.
1"con_01M0E4J0BBE1FVDVTZ1BSS5C97"
Optional display name for the run.
"Appointment booking release check"
Optional exact list of the suite's test IDs and current version IDs. Each test and version must appear once. The request is refused if the suite membership or any version changed. Omit this field to use the current suite.
Show child attributes
Show child attributes
Response
The run header for a new request or an idempotent replay. The run may still be executing or grading.
111Simulation execution status. A completed run can contain failed simulations, and grading may still be in progress.
pending, running, completed, canceled 11voice, chat Number of test-and-persona combinations captured when the run started.
Show child attributes
Show child attributes
Simulations whose execution completed, failed, or was canceled.
Simulations eligible for grading under the run's frozen grader selection.
Gradable simulations whose grading is complete or errored. This is not a count of passed simulations.
Open this URL in a browser to follow the run and inspect its results.