> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bland.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Get Check Run

> Retrieve a check run and its verdicts.

### Overview

Returns one check run with its per-judge verdicts. Poll this after [Start Check Run](/api-v2/post/agents-id-environments-env-check-runs) until `status` is `PASSED`, `FAILED`, `ERROR`, or `CANCELLED`; only then are `overall_passed` and `verdicts` final. The run is a report only: it never changes which version an environment runs.

### Headers

<ParamField header="authorization" type="string" required>
  Your API key for authentication.
</ParamField>

### Path Parameters

<ParamField path="agent_id" type="string" required>
  The agent's unique identifier.
</ParamField>

<ParamField path="run_id" type="string" required>
  The check run's unique identifier, from the `id` returned by [Start Check Run](/api-v2/post/agents-id-environments-env-check-runs).
</ParamField>

### Response

<ResponseField name="data" type="object">
  The check run.

  <Expandable title="check run object">
    <ResponseField name="id" type="string">
      Unique identifier for the check run.
    </ResponseField>

    <ResponseField name="agent_id" type="string">
      The agent the run belongs to.
    </ResponseField>

    <ResponseField name="env_type" type="string">
      The environment whose checks were run: `staging` or `production`.
    </ResponseField>

    <ResponseField name="agent_version_id" type="string">
      The candidate version the run scored.
    </ResponseField>

    <ResponseField name="semver" type="string | null">
      The candidate's published version number. `null` when the candidate has never been published, which is the normal case for a `staging` run.
    </ResponseField>

    <ResponseField name="version_name" type="string | null">
      The candidate's saved name, or `null` if it has none.
    </ResponseField>

    <ResponseField name="status" type="string">
      `PENDING` (created, scoring not yet started), `RUNNING` (generating and scoring conversations), then one of the terminal states `PASSED`, `FAILED`, `ERROR`, or `CANCELLED`. A run that stays `PENDING` or `RUNNING` for 45 minutes is marked `ERROR` with the message `Check run timed out`.
    </ResponseField>

    <ResponseField name="judge_names" type="array">
      Display names of the judges frozen into the run when it started. Present while the run is in flight, before any verdicts exist.
    </ResponseField>

    <ResponseField name="scenario_count" type="integer">
      Number of scenarios the run drove conversations from.
    </ResponseField>

    <ResponseField name="simulations_count" type="integer">
      Conversations generated per scenario.
    </ResponseField>

    <ResponseField name="overall_passed" type="boolean | null">
      `true` when the run `PASSED`, meaning every judge with `required: true` passed. `false` when it `FAILED`. `null` while in flight, or when the run ended in `ERROR` or `CANCELLED` without a verdict. Informational judges (`required: false`) never affect this value.
    </ResponseField>

    <ResponseField name="verdicts" type="array | null">
      Per-judge results, or `null` until the run completes. See `data.verdicts` below.
    </ResponseField>

    <ResponseField name="simulation_set_id" type="string | null">
      ID of the generated conversation set. `null` until generation starts.
    </ResponseField>

    <ResponseField name="eval_run_id" type="string | null">
      ID of the scoring run. `null` until scoring starts.
    </ResponseField>

    <ResponseField name="error_message" type="string | null">
      Why the run ended in `ERROR`, or `null`.
    </ResponseField>

    <ResponseField name="triggered_by" type="string | null">
      ID of the user who started the run, or `null` when started with an org-level key.
    </ResponseField>

    <ResponseField name="created_at" type="string">
      ISO 8601 timestamp the run was created.
    </ResponseField>

    <ResponseField name="started_at" type="string | null">
      When scoring began. `null` while `PENDING`.
    </ResponseField>

    <ResponseField name="completed_at" type="string | null">
      When the run reached a terminal status. `null` while in flight.
    </ResponseField>
  </Expandable>
</ResponseField>

<ResponseField name="data.verdicts" type="array | null">
  One entry per judge, in config order. `null` until the run completes.

  <Expandable title="verdict object">
    <ResponseField name="eval_agent_id" type="string">
      The judge's unique identifier.
    </ResponseField>

    <ResponseField name="eval_agent_version_id" type="string">
      The judge version the run used.
    </ResponseField>

    <ResponseField name="name" type="string">
      The judge's display name.
    </ResponseField>

    <ResponseField name="required" type="boolean">
      Whether a failing verdict from this judge fails the run.
    </ResponseField>

    <ResponseField name="target_level_keys" type="array">
      The grade levels counted as passing. Empty for pass/fail judges.
    </ResponseField>

    <ResponseField name="match_rate" type="number | null">
      Fraction of scored conversations, `0` to `1`, that landed on a passing level. `null` when no conversation produced a determinable result.
    </ResponseField>

    <ResponseField name="score" type="number | null">
      Mean judge score across scored conversations on a `0` to `100` scale, rounded to one decimal. `null` when none were scored.
    </ResponseField>

    <ResponseField name="passed" type="boolean | null">
      `true` when `match_rate` is at least `0.5`, `false` when it is lower, `null` when `match_rate` is `null`.
    </ResponseField>
  </Expandable>
</ResponseField>

<ResponseField name="errors" type="null | array">
  `null` on success, or a list of error objects if the request failed.
</ResponseField>

<ResponseExample>
  ```json Response theme={null}
  {
    "data": {
      "id": "fb12d904-efbe-4bb2-b167-04a5d8446fad",
      "agent_id": "3a5dff19-707a-4776-943e-809520a3592a",
      "env_type": "production",
      "agent_version_id": "20752a9f-c33c-49b1-914c-ba468ea4ad5f",
      "semver": "2.1.0",
      "version_name": "Insurance intake v2",
      "status": "PASSED",
      "judge_names": ["Resolution quality", "No policy violations"],
      "scenario_count": 2,
      "simulations_count": 5,
      "overall_passed": true,
      "verdicts": [
        {
          "eval_agent_id": "30dd62b6-13b2-4279-ac57-fdee065880a5",
          "eval_agent_version_id": "2ea22a1c-380f-4d94-90ce-0c30b003fbfc",
          "name": "Resolution quality",
          "required": true,
          "target_level_keys": ["good", "excellent"],
          "match_rate": 0.8,
          "score": 81.3,
          "passed": true
        },
        {
          "eval_agent_id": "0497636a-27e3-4b06-b05a-3f49a6610f4f",
          "eval_agent_version_id": "aed97f7b-4535-49d3-b125-c4692ab44d87",
          "name": "No policy violations",
          "required": false,
          "target_level_keys": [],
          "match_rate": 0.4,
          "score": 40.0,
          "passed": false
        }
      ],
      "simulation_set_id": "92e38d39-f377-4075-9ddc-724096d72d22",
      "eval_run_id": "4f96db0e-2456-4b70-81e9-155b3c29a9b5",
      "error_message": null,
      "triggered_by": "0ce6e187-38fa-40f7-95df-465368d7c104",
      "created_at": "2026-09-10T14:03:22.901Z",
      "started_at": "2026-09-10T14:03:25.144Z",
      "completed_at": "2026-09-10T14:16:58.377Z"
    },
    "errors": null
  }
  ```

  ```json Not Found theme={null}
  {
    "data": null,
    "errors": [
      {
        "error": "NOT_FOUND",
        "message": "Check run not found"
      }
    ]
  }
  ```
</ResponseExample>

***

Docs for agents: [llms.txt](/llms.txt)