---
title: "Extract clean page content"
method: POST
path: "/extract"
---

# Extract clean page content

`POST /extract`

## Request

- Base URL: `https://search.neurobird.com`
- URL: `https://search.neurobird.com/extract`
- Auth: none required

Fetch one or more URLs and return the main content as clean markdown, with navigation, adverts and boilerplate removed. PDFs are converted to text.

Pass `query` to trim each page to the passages relevant to it. Pass `schema` to fill a JSON Schema from each page. Costs 1 credit per page.

## Request body

- ExtractRequest
  - `urls` string[], required
  - `format` 'markdown' | 'text'
  - `include_images` boolean
  - `schema` object, nullable
  - `query` string, nullable

## Response `200`

Successful Response

- ExtractResponse
  - `results` ExtractResult[], required
    - `url` string, required
    - `title` string
    - `raw_content` string
    - `images` string[]
    - `structured` object, nullable
    - `words` integer
  - `failed_results` ExtractFailure[]
    - `url` string, required
    - `error` string, required
  - `response_time` number, required
  - `request_id` string, required
  - `credits` integer

## Other responses

- `422`: Validation Error

---

[API](https://skmtc.dev/neurobird/apis/neurobird-search-api.md) · [All operations](https://skmtc.dev/neurobird/apis/neurobird-search-api/llms.txt) · [OpenAPI document](https://skmtc.dev/neurobird/apis/neurobird-search-api/revisions/103dcd55a8f5?raw)