[← Back to tools]
firecrawl.extract

Extract structured data with Firecrawl: up to $0.01 per call, no Firecrawl account

Use Firecrawl from Claude, Claude Code, Cursor, ChatGPT, and other agents through one Cralo balance.

Extract structured data from one public page with Firecrawl using a caller-supplied JSON schema. Billed from Firecrawl extract credits, up to $0.10 per call.

Price and execution

Price
up to $0.01
Pricing model
Up to this amount, billed on success
Execution class
fast

Limits

No additional limits are published for this tool.

Parameters

Fields firecrawl.extract accepts as JSON input.

  • urlstring (uri)Required

    Public page URL to extract structured data from.

  • promptstringOptional

    Optional instructions that guide the extraction.

  • schemaobjectRequired

    A JSON Schema describing the object Firecrawl should extract.

  • enable_web_searchbooleanOptional

    When true, Firecrawl may use web search to supplement extraction.

Raw input JSON Schema
{
  "type": "object",
  "required": [
    "url",
    "schema"
  ],
  "properties": {
    "url": {
      "type": "string",
      "format": "uri",
      "pattern": "^https?://",
      "description": "Public page URL to extract structured data from."
    },
    "prompt": {
      "type": "string",
      "maxLength": 2000,
      "minLength": 1,
      "description": "Optional instructions that guide the extraction."
    },
    "schema": {
      "type": "object",
      "minProperties": 1,
      "description": "A JSON Schema describing the object Firecrawl should extract."
    },
    "enable_web_search": {
      "type": "boolean",
      "default": false,
      "description": "When true, Firecrawl may use web search to supplement extraction."
    }
  },
  "additionalProperties": false
}

Returns

Fields a successful firecrawl.extract call returns.

  • dataobjectRequired
  • titlestring | nullRequired
  • source_urlstringRequired
  • retrieved_atstring (date-time)Required
Raw output JSON Schema
{
  "type": "object",
  "required": [
    "data",
    "source_url",
    "title",
    "retrieved_at"
  ],
  "properties": {
    "data": {
      "type": "object"
    },
    "title": {
      "type": [
        "string",
        "null"
      ]
    },
    "source_url": {
      "type": "string"
    },
    "retrieved_at": {
      "type": "string",
      "format": "date-time"
    }
  },
  "additionalProperties": false
}

Run this tool

Copy this instruction to your agent to set up Cralo and call this tool with the example input.

Set up Cralo by following https://cralo.ai/SKILL.md, then call the firecrawl.extract tool with this input: {"url":"https://example.com","schema":{"type":"object","properties":{"title":{"type":"string"}}},"prompt":"Extract the page title"}

Safety notes

Side Effect
read_only
Public Sources Only
Yes

How this compares to using Firecrawl directly

You do not need a separate Firecrawl account or subscription for this call. Cralo uses one prepaid balance across all tools and shows the per-call price before you run the command.

Related tools

[Browse all tools]