Skip to main content
POST
/
evaluations
/
runs:runAgent
{
  "id": "<string>",
  "status": "<string>",
  "error": "<string>",
  "result": [
    {
      "evaluator": "<string>",
      "evaluatorId": "<string>",
      "score": 123,
      "status": "<string>",
      "reason": "<string>",
      "version": "<string>",
      "threadId": "<string>",
      "runId": "<string>",
      "error": "<string>",
      "additionalDetails": {}
    }
  ]
}

Authorizations

​
Authorization
string
header
required

The access token received from the authorization server in the OAuth 2.0 flow.

Query Parameters

​
api-version
string
required

The API version to use for this operation.

Minimum length: 1

Body

application/json

Agent evaluation to be run

Evaluation request for agent run.

​
runId
string
required

Identifier of the agent run.

​
evaluators
object
required

Evaluators to be used for the evaluation.

​
appInsightsConnectionString
string
required

Pass the AppInsights connection string to the agent evaluation for the evaluation results and the errors logs.

​
threadId
string

Identifier of the agent thread. This field is mandatory currently, but it will be optional in the future.

​
samplingConfiguration
object

Definition for sampling strategy.

​
redactionConfiguration
object

The redaction configuration will allow the user to control what is redacted.

Response

The request has succeeded and a new resource has been created as a result.

Evaluation response for agent evaluation run.

​
id
string
required

Identifier of the agent evaluation run.

​
status
string
required

Status of the agent evaluation. Options: Running, Completed, Failed.

​
error
string

The reason of the request failure for the long running process, if applicable.

​
result
object[]

The agent evaluation result.