Model Context Protocol

ScrapingBot MCP Server

Give Claude, Cursor and other AI agents the ability to scrape any website — without leaving the conversation. One config block, zero maintenance.

Get your API key →API Documentation

What is MCP?

Model Context Protocol (MCP) is an open standard from Anthropic that lets AI assistants call external tools — APIs, databases, file systems — directly from the conversation.

Why ScrapingBot?

Ask Claude to research a competitor, compare prices, or monitor a website. It calls ScrapingBot automatically. No code to write, no proxy to manage.

Works everywhere

Compatible with Claude Desktop, Claude Code, Cursor, Windsurf, and any IDE or agent framework that supports MCP.

Setup in 2 minutes

No install required — the server runs on demand via npx.

Get your credentials

Sign up at app.scraping-bot.io — 1,000 free credits/month, no credit card required. Find your username and API key in the dashboard.

Add the MCP server config

Add this to your claude_desktop_config.json file. On macOS: ~/Library/Application Support/Claude/. On Windows: %APPDATA%\Claude\.

{
  "mcpServers": {
    "scraping-bot": {
      "command": "npx",
      "args": ["-y", "@scraping-bot/mcp-server"],
      "env": {
        "SCRAPING_BOT_USERNAME": "your_username",
        "SCRAPING_BOT_API_KEY": "your_api_key"
      }
    }
  }
}

Replace your_username and your_api_key with your credentials from app.scraping-bot.io.

Restart your AI client and start scraping

ScrapingBot tools appear automatically. Ask Claude anything that requires live web data.

Available tools

6 tools covering all ScrapingBot endpoints. Claude picks the right one automatically based on the URL.

Tool	Description	Parameters
scrape_raw_html	Raw HTML of any page, with optional JS rendering	url, useChrome?, premiumProxy?
scrape_retail	Structured product data from e-commerce pages	url, premiumProxy?
scrape_real_estate	Real estate listings (price, surface, location…)	url, premiumProxy?
scrape_search_engine	Google / Bing SERP results in structured JSON	url, premiumProxy?
scrape_social_media	Public social media profiles and posts	url, premiumProxy?
get_usage	Check remaining API credits	—

Example prompt

User: "Scrape the top 5 Google results for 'best web scraping API 2025'
        and compare their pricing pages."

Claude: I'll search Google and scrape each pricing page for you.
[calls scrape_search_engine with url="https://www.google.com/search?q=best+web+scraping+API+2025"]
[calls scrape_raw_html for each of the 5 result URLs]

Here's the comparison: ...

Claude calls scrape_search_engine and scrape_raw_html automatically — no code, no copy-pasting URLs.

Ready to give your AI real-time web access?

1,000 free scraping credits every month. No credit card required.

Start for free →