Hybrid Vision Spider | AI-Powered Universal Web Scraper

Point. Click. Extract. Data from any webpage in seconds.

---

What This Actor Does

This actor visits web pages and extracts the data you need. Just select what you want to extract using simple checkboxes - no coding required.

Works great for:

• Contact information (emails, phone numbers)
• Product data (names, prices, descriptions)
• Social media links
• Any structured data from websites

Quick Start (3 Steps)

Step 1: Enter URLs

Paste the URLs you want to scrape, one per line:

https://example.com/contact
https://example.com/about
https://example.com/products

Step 2: Select What to Extract

Check the boxes for the data you want:

Checkbox	What It Extracts
Extract Page Title	Main title/heading
Extract Email	Email addresses found on page
Extract Phone	Phone numbers
Extract Links	Important links
Extract Images	Image URLs
Extract Price	Price values ($, EUR, etc.)
Extract Social Links	Facebook, Twitter, LinkedIn, etc.
Extract Address	Physical addresses
Extract Description	Page summary/meta description

Step 3: Choose Mode and Run

Mode	When to Use	Cost
HTML Only	Simple pages, contact info	Free
Hybrid	Most websites (recommended)	Low
Vision Only	Complex pages, visual data	Higher

Click Start and wait for results!

Extraction Modes Explained

HTML Only Mode (Free)

• Uses pattern matching to find data
• Works offline, no AI costs
• Best for: emails, phones, links, prices
• Limitation: Cannot understand page context

Hybrid Mode (Recommended)

• Tries HTML first, uses AI if needed
• Best balance of speed and accuracy
• Requires: OpenAI API key
• Cost: ~$0.001-0.005 per page

Vision Only Mode

• AI "sees" the page like a human
• Can extract any visible data
• Requires: OpenAI API key
• Cost: ~$0.002-0.010 per page

OpenAI API Key

Required for Hybrid and Vision Only modes.

How to get one:

1. Go to platform.openai.com/api-keys
2. Create an account or sign in
3. Click "Create new secret key"
4. Copy the key (starts with sk-)
5. Paste it in the "OpenAI API Key" field

Advanced: Custom JSON Schema

For complex extractions, use the "Custom Schema" field. This overrides the checkboxes.

Example - Extract product data:

{
  "type": "object",
  "properties": {
    "productName": {
      "type": "string",
      "description": "Name of the product"
    },
    "price": {
      "type": "number",
      "description": "Price in USD"
    },
    "rating": {
      "type": "number",
      "description": "Rating out of 5 stars"
    },
    "reviews": {
      "type": "array",
      "items": { "type": "string" },
      "description": "Customer reviews"
    }
  },
  "required": ["productName", "price"]
}

Tips:

• Add description to help the AI understand what to look for
• Use required for must-have fields
• Use type: "array" for lists of items

Output Format

Each result contains:

{
  "url": "https://example.com/contact",
  "data": {
    "title": "Contact Us",
    "email": "hello@example.com",
    "phone": "+1 555-123-4567"
  },
  "confidence": {
    "title": 0.95,
    "email": 0.90,
    "phone": 0.85
  },
  "timestamp": "2024-12-14T10:30:00.000Z"
}

Cost Control

Set limits to control your spending:

Setting	What It Does	Recommended
Max Results	Stop after X items	100
Max AI Pages	Limit AI-processed pages	10
Token Budget	Cap total AI tokens	50,000

Estimated costs:

• HTML Only: Free
• Hybrid (10 pages): ~$0.01-0.05
• Vision Only (10 pages): ~$0.02-0.10

Limitations

• Requires OpenAI API key for AI modes
• Cannot follow links (processes only provided URLs)
• Some JavaScript-heavy sites may need Vision mode
• Rate limited by OpenAI API

---

Built with 🩶 for the Apify community 🫡