Firecrawl: Extract

Extract structured data from web page URLs or websites with an AI prompt and a JSON schema, as an asynchronous job.

Tool ID: firecrawl_extract

Run this action

Use the TypeScript SDK for one call. Use ctx.tools.execute(...) inside a Play for durable calls, scheduled calls, or CSV processing.

import { Deepline } from 'deepline';

const deepline = await Deepline.connect();
const result = await deepline.tools.execute(
  'firecrawl_extract',
  {
    "urls": [
      "https://example.com"
    ]
  },
);

console.log(result.toolResponse.raw);

CLI

deepline tools execute firecrawl_extract --input '{
  "urls": [
    "https://example.com"
  ]
}' --json

Example response

The SDK returns this shape at result.toolResponse.raw. The example values illustrate the response.

{
  "data": {
    "status": "example"
  }
}

Use deepline describe firecrawl_extract --json for the latest machine-readable contract.

Input reference

Start an extraction job and return its ID for status polling.

Name	Type	Required	Default	Details
`payload.urls`	`array`	Yes	—	URLs to extract data from
`payload.prompt`	`string`	No	—	Natural language extraction instructions
`payload.schema`	`record`	No	—	JSON schema for structured output
`payload.enableWebSearch`	`boolean`	No	—	Enable web search to augment extraction
`payload.ignoreSitemap`	`boolean`	No	—	Ignore sitemap when discovering pages (default false)
`payload.includeSubdomains`	`boolean`	No	—	Include subdomains in extraction (default true)
`payload.showSources`	`boolean`	No	—	Include source URLs in response (default false)
`payload.ignoreInvalidURLs`	`boolean`	No	—	Skip invalid URLs instead of failing (default true)
`payload.scrapeOptions`	`object`	No	—	Scrape options for each URL
`payload.timeoutMs`	`integer`	No	—	Deprecated compatibility field. Start actions now return immediately. Use the matching status action to poll.

This input schema is too large for an inline display. Read the complete live contract with deepline describe firecrawl_extract --json.

Output reference

Structured extraction result.

Name	Type	Required	Default	Details
`result.data`	`object`	Yes	—	Provider response payload.
`result.data.id`	`string`	No	—	—
`result.data.status`	`string`	Yes	—	—
`result.data.data`	`unknown`	No	—	—
`result.data.tokensUsed`	`number`	No	—	—
`result.data.expiresAt`	`string`	No	—	—
`result.meta`	`record`	No	—	Additional response metadata (status, paging).

Deepline cost

Pricing model: provider_usage (provider usage).
Estimated Deepline credits: 0 per pricing unit.

​Run this action

​CLI

​Example response

​Input reference

​Output reference

​Deepline cost

​Related documentation

Run this action

CLI

Example response

Input reference

Output reference

Deepline cost

Related documentation