AutoScraper API Reference

The enhanced LLM Extraction API now supports user-defined JSON Schema to enforce strict data structure compliance while retaining natural language flexibility. Combine free-text instructions with schema validation to balance precision and adaptability, ideal for regulated industries (e.g., finance, healthcare) or complex nested data.

Scrape content from a URL

This endpoint allows you to scrape markdown content from a given URL. You must provide a valid URL as a query parameter.

Required parameters

Name
url
Type
string
Description
The URL from which the content will be scraped. For example: https://www.maginative.com/
Name
prompt
Type
string
Description
Natural language extraction prompt (≤500 chars).
Name
schema
Type
string
Description
A JSON Schema defining the expected output structure. LLM will prioritize filling this schema over free-text interpretation.
Name
proxy_premium
Type
boolean
Description
Use premium proxies to make the request harder to detect
Name
proxy_country
Type
string
Description
Geolocation of the IP used to make the request. Only for Premium Proxies, ISO 3166 country codes.

Request

POST

/v1/scraper/schema

curl -G https://api.autoscraper.pro/v1/scraper/schema \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
      "urls": [
        "https://www.example.com/"
      ],
      "proxy_premium": true,
      "proxy_country": ""
    }'

Response

{
  "product_name": "Wireless Noise-Canceling Headphones",
  "brand": "AudioTech",
  "pricing": {
    "base_price": 299.99,
    "discount": 0.15
  },
  "specifications": [
    "Battery life: 30hrs",
    "Bluetooth 5.3"
  ]
}