Extract (Structured Data)

Extract structured data from one or more web pages using AI. Scrapes the provided URLs, combines content, and uses an LLM to return JSON matching your schema.

POST/v1/extract

Scrape URLs and extract structured data using an LLM with schema-constrained output.

Request Body

Parameter	Type	Description
`urls`required	string[]	URLs to extract data from (1-10)
`schema`required	Record<string, any>	JSON Schema describing the desired output structure
`prompt`required	string	Natural language extraction instruction (e.g., "Extract product name, price, and rating")
`type`	"html" \| "markdown"	Content format fed to the LLM. Defaults to "markdown"
`proxy`	{ country: string }	Proxy for geo-targeted requests. See supported countries

Example Request

cURL

Node.js

curl -X POST https://api.builde2e.com/api/v1/extract \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer uc-YOUR-API-KEY' \
  -d '{
    "urls": ["https://example.com/product/laptop-x1"],
    "schema": {
      "type": "object",
      "properties": {
        "name": { "type": "string" },
        "price": { "type": "number" },
        "features": { "type": "array", "items": { "type": "string" } }
      }
    },
    "prompt": "Extract the product details from this page"
  }'

Example Response

json

{
  "success": true,
  "data": {
    "products": [
      { "name": "Widget Pro", "price": 29.99, "rating": 4.5 },
      { "name": "Widget Max", "price": 49.99, "rating": 4.8 }
    ]
  },
  "sources": ["https://example.com/product-1", "https://example.com/product-2"],
  "cost": 0.009,
  "timestamp": "2026-03-04T10:15:30.123Z"
}

Response Structure

Field	Type	Description
`success`	boolean	Whether extraction succeeded
`data`	any	Extracted structured data matching your schema. null on failure.
`sources`	string[]	URLs that were successfully scraped and used for extraction
`cost`	number	Cost in USD ($0.002 per successfully scraped URL + $0.005 per extraction)
`error`	string	Error message if extraction failed
`timestamp`	string	ISO timestamp of the response

Using Zod Schemas — SDK Only

The Node.js SDK supports Zod schemas for type-safe extraction. Pass a Zod schema instead of JSON Schema and get fully typed results.

Node.js

import { z } from 'zod';

const ProductSchema = z.object({
  name: z.string(),
  price: z.number(),
  currency: z.string(),
  features: z.array(z.string()),
  inStock: z.boolean()
});

const result = await BuildE2E.extract({
  urls: ['https://example.com/product/laptop-x1'],
  schema: ProductSchema,
  prompt: 'Extract the product details'
});

// result.data is typed as z.infer<typeof ProductSchema>
console.log(result.data.name);
console.log(result.data.price);

Pricing

Extraction is charged in two parts:

Component	Cost (USD)	Description
Per URL scraped	`$0.002`	Charged per successfully scraped URL (failed URLs are free)
Per extraction	`$0.005`	One-time LLM extraction fee per request

Example: extracting from 2 URLs costs 2 x $$0.002 + $$0.005 = $$0.009.