Skip to main content
POST
/
scrape
Scrape Content
curl --request POST \
  --url https://api.example.com/scrape \
  --header 'Content-Type: application/json' \
  --data '
{
  "url": "<string>",
  "async": true,
  "render": true,
  "method": "<string>",
  "format": "<string>",
  "country": "<string>",
  "includeHeaders": true,
  "headers": {},
  "body": {},
  "extract": {},
  "extract.schema": {},
  "extract.prompt": "<string>",
  "extract.systemPrompt": "<string>",
  "extract.model": "<string>",
  "extract.includeMetadata": true
}
'

​
Overview

The /scrape endpoint initiates a new web scraping job with the specified URL and options. It returns the scraped content directly and supports both renderless and renderful scraping modes. You can also use AI-powered extraction to get structured data from the page.

​
Parameters

​
Request Body

​
url
string
required
The URL to scrape
​
async
boolean
default:"false"
Whether the request should be asynchronous. When true, returns a job ID immediately that can be used to check status.
​
render
boolean
default:"false"
Whether to render JavaScript on the page. Enable for dynamic content loaded via JavaScript.
​
method
string
default:"get"
HTTP method for the request. Options: get, post, put, delete, patch, head, options
​
format
string
default:"raw"
Response format. Options: raw, json, markdown
​
country
string
default:"us"
Proxy country for geo-targeted requests (e.g., us, uk, de)
​
includeHeaders
boolean
default:"false"
Whether to include response headers in the response
​
headers
object
HTTP headers for the request
{
  "User-Agent": "Mozilla/5.0...",
  "Accept-Language": "en-US,en;q=0.9"
}
​
body
object
Request body content (for POST/PUT requests)

​
LLM Extraction Options

​
extract
object
AI-powered extraction options. When provided, the scraped content will be processed by an LLM to extract structured data.
​
extract.schema
object
JSON Schema defining the structure to extract. Use this for precise, typed extraction.
{
  "type": "object",
  "properties": {
    "title": { "type": "string", "description": "The page title" },
    "price": { "type": "number", "description": "Product price" },
    "features": {
      "type": "array",
      "items": { "type": "string" }
    }
  },
  "required": ["title", "price"]
}
​
extract.prompt
string
Natural language prompt describing what to extract. Use this for flexible, conversational extraction.
​
extract.systemPrompt
string
Custom system prompt to guide the LLM behavior
​
extract.model
string
default:"gpt-4o-mini"
LLM model to use for extraction. Options: gpt-4o, gpt-4o-mini, claude-3-5-sonnet
​
extract.includeMetadata
boolean
default:"true"
Whether to include extraction metadata (tokens used, cost) in response

​
Example Requests

​
Basic Scraping

curl -X POST "https://api.scrapengine.io/api/v1/scrape" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.example.com/product",
    "render": true
  }'

​
With LLM Extraction (Schema)

Extract structured data using a JSON schema: 
curl -X POST "https://api.scrapengine.io/api/v1/scrape" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.example.com/product",
    "render": true,
    "extract": {
      "schema": {
        "type": "object",
        "properties": {
          "productName": { "type": "string", "description": "The product name" },
          "price": { "type": "number", "description": "Price in USD" },
          "rating": { "type": "number", "description": "Average rating out of 5" },
          "features": {
            "type": "array",
            "items": { "type": "string" },
            "description": "List of product features"
          }
        },
        "required": ["productName", "price"]
      },
      "model": "gpt-4o-mini"
    }
  }'

​
With LLM Extraction (Prompt)

Extract data using a natural language prompt: 
curl -X POST "https://api.scrapengine.io/api/v1/scrape" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://www.example.com/article",
    "extract": {
      "prompt": "Extract the article title, author name, publication date, and a brief summary of the main points",
      "model": "claude-3-5-sonnet"
    }
  }'

​
Response

​
Success Response (200)

Without extraction - Returns the scraped HTML content directly: 
<html>
  <head>
    <title>Example Product</title>
  </head>
  <body>
    <!-- Scraped content here -->
  </body>
</html>
With extraction - Returns structured JSON: 
{
  "data": {
    "productName": "Premium Wireless Headphones",
    "price": 299.99,
    "rating": 4.5,
    "features": [
      "Active Noise Cancellation",
      "40-hour battery life",
      "Bluetooth 5.0"
    ]
  },
  "metadata": {
    "tokensUsed": 1250,
    "model": "gpt-4o-mini"
  }
}

​
Response Headers

HeaderDescription
x-remaining-creditsNumber of API credits remaining
x-trace-idUnique identifier for the request

​
Error Responses

StatusDescription
400Bad Request - Invalid parameters or URL
401Unauthorized - Invalid or missing API key
403Forbidden - Access denied to target resource
404Not Found - Target URL not found
408Request Timeout
429Too Many Requests - Rate limit exceeded
500Internal Server Error
550Faulted After Retries - Job failed after multiple attempts
Error Response Format: 
{
  "error": "Error description message",
  "traceId": "abc123-def456",
  "timestamp": "2024-01-15T10:30:00.000Z"
}

​
Use Cases

  • E-commerce scraping: Extract product information, prices, and availability
  • Content aggregation: Collect articles, blog posts, and news content
  • Lead generation: Extract contact information and company details
  • Competitor analysis: Monitor competitor websites and pricing
  • SEO analysis: Extract meta tags, headings, and content structure
  • AI-powered extraction: Use LLM to extract structured data without writing parsers