curl --request POST \
--url https://app.egma.ai/v1/tests \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"suiteId": "ste_01M0E4EVJ6ECGVJEA4NSBTC0CC",
"name": "No available appointments",
"scenario": "Ask Harbor Clinic for an afternoon appointment. If none is available, ask how to arrange a callback.",
"expectedBehaviors": [
"The agent calls check_availability.",
"The agent says that no appointments are available and does not offer a time."
],
"personas": [
"Everyday caller"
]
}
'{
"id": "<string>",
"projectId": "<string>",
"suiteId": "<string>",
"name": "<string>",
"description": "<string>",
"version": 2,
"versionId": "<string>",
"scenario": "<string>",
"expectedBehaviors": [
"<string>"
],
"personas": [
{
"id": "<string>",
"name": "<string>",
"archivedAt": "2023-11-07T05:31:56Z"
}
],
"mockTools": [
{
"tool": "<string>",
"answer": "<unknown>",
"error": "<unknown>"
}
],
"env": {
"retell_dynamic_variables": {},
"job_dispatch_metadata": {}
},
"revision": "<string>",
"createdAt": "2023-11-07T05:31:56Z",
"updatedAt": "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": {}
}Create a test in a test suite
Creates the test and its first content version in an existing suite. Choose at least one persona. The agent and connection are selected when starting a run, not when creating the test.
curl --request POST \
--url https://app.egma.ai/v1/tests \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"suiteId": "ste_01M0E4EVJ6ECGVJEA4NSBTC0CC",
"name": "No available appointments",
"scenario": "Ask Harbor Clinic for an afternoon appointment. If none is available, ask how to arrange a callback.",
"expectedBehaviors": [
"The agent calls check_availability.",
"The agent says that no appointments are available and does not offer a time."
],
"personas": [
"Everyday caller"
]
}
'{
"id": "<string>",
"projectId": "<string>",
"suiteId": "<string>",
"name": "<string>",
"description": "<string>",
"version": 2,
"versionId": "<string>",
"scenario": "<string>",
"expectedBehaviors": [
"<string>"
],
"personas": [
{
"id": "<string>",
"name": "<string>",
"archivedAt": "2023-11-07T05:31:56Z"
}
],
"mockTools": [
{
"tool": "<string>",
"answer": "<unknown>",
"error": "<unknown>"
}
],
"env": {
"retell_dynamic_variables": {},
"job_dispatch_metadata": {}
},
"revision": "<string>",
"createdAt": "2023-11-07T05:31:56Z",
"updatedAt": "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": {}
}Authorizations
An Egma API key.
Query Parameters
1Body
An existing active test suite in the current project.
1The situation and goal the synthetic caller acts out.
Statements checked independently against the completed conversation. Include at least one non-empty statement.
At least one available persona ID or unambiguous current name. Each selected persona creates one simulation for this test in a run.
Test-owned tool answers. Use an empty array to remove existing mocks on update. Applying mocks requires a supported connection and, for LiveKit, tool instrumentation.
One named tool's fixed response for this test. Supply exactly one of answer or error. Tools without a mock run normally.
- Option 1
- Option 2
Show child attributes
Show child attributes
Non-secret startup context for the agent's provider. Omit to keep existing context on update, or send null to clear it.
Show child attributes
Show child attributes
Response
The new test.
111x >= 11Show child attributes
Show child attributes
One named tool's fixed response for this test. Supply exactly one of answer or error. Tools without a mock run normally.
- Option 1
- Option 2
Show child attributes
Show child attributes
Show child attributes
Show child attributes
1