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

# Agent Extraction

> Agent Extraction is discontinued on 18 September 2026 with no automatic replacement — recreate the fields you rely on as your own extractions.

<Warning>
  **`agent_extraction` → Discontinued — 18 September 2026.** `agent_extraction` returns `null`. The key stays in the payload.
</Warning>

Agent Extraction runs on templates Bolna maintains rather than ones you configure. There is no replacement template and no migration performed on your behalf.

<Warning>
  **No automatic path.** To keep equivalent data after 18 September 2026, recreate the fields you rely on as your own extractions. The template's internal prompt is not exposed, so you are writing new extractions rather than copying existing ones.
</Warning>

<Note>
  **Auto-reschedule is unaffected.** Automatic rescheduling of callbacks continues to work after 18 September 2026 with no action from you. It is handled separately from Agent Extraction.
</Note>

## What to do

The template id (`asst_…`) is not something you can look up, so identify the fields from a recent payload.

<Steps>
  <Step title="Pull a completed execution">
    Read the object under `agent_extraction`.
  </Step>

  <Step title="Note every key inside it">
    Those are the fields the template produced.
  </Step>

  <Step title="Create a category and one extraction per field you need">
    Follow [Extraction](/docs/migrating-to-extractions/extraction). Set each name to the old field name to keep result keys stable.
  </Step>

  <Step title="Write your own question for each">
    State what to look at, what to ignore, and what each pre-defined option means.
  </Step>

  <Step title="Test against a real past transcript before the cutoff">
    See [Verifying a migration](/docs/migrating-to-extractions/overview#verifying-a-migration).
  </Step>
</Steps>

## Before and after

<CodeGroup>
  ```json Before: keyed by template id theme={"system"}
  {
    "agent_extraction": {
      "asst_0DxYqjpXy2EObi8cKPs7fwRN": {
        "Final Evaluation Rating": "Strong Fit"
      }
    }
  }
  ```

  ```json After: a category you created and named theme={"system"}
  {
    "agent_extraction": null,
    "extracted_data": {
      "Candidate Evaluation": {
        "Final Evaluation Rating": {
          "objective": "Strong Fit",
          "confidence": 0.86, …
        }
      }
    }
  }
  ```
</CodeGroup>

The opaque `asst_` key is replaced by a category name you control, which is easier to read. But the values behind it come from your prompt, not the template's.

<Warning>
  **Results will not match exactly.** You are authoring new questions against the same transcripts, so scores and wording will shift. If Agent Extraction output feeds scoring, hiring decisions, reporting, or anything with an audit trail, revalidate against past calls rather than assuming parity. Reports spanning 18 September will show a discontinuity that is not caused by a change in call quality.
</Warning>

## Other features

<CardGroup cols={3}>
  <Card title="Extraction" icon="list-check" href="/docs/migrating-to-extractions/extraction">
    `extracted_data` · flat keys
  </Card>

  <Card title="Custom Analytics" icon="sliders" href="/docs/migrating-to-extractions/custom-analytics">
    `custom_extractions`
  </Card>

  <Card title="Call Summary" icon="file-lines" href="/docs/migrating-to-extractions/call-summary">
    `summary`
  </Card>
</CardGroup>

Email [support@bolna.ai](mailto:support@bolna.ai) to remove a legacy summarisation task, or with an `execution_id` for a payload that does not match this guide.

## Next Steps

<CardGroup cols={3}>
  <Card title="Using Extractions" icon="list-check" href="/docs/prompting/using-extractions">
    Configure extractions in the dashboard
  </Card>

  <Card title="Dispositions API" icon="code" href="/docs/api-reference/dispositions/overview">
    Manage extractions programmatically
  </Card>

  <Card title="Webhooks" icon="webhook" href="/docs/post-call/polling-call-status-webhooks">
    Receive extraction data in real-time
  </Card>
</CardGroup>