Crawlbase MCP Server: Make Your LLMs Web-Aware in Seconds ✨
Connect Claude, Cursor, Windsurf, and other MCP-compatible AI Agents to live web data — with just one config.
🎁 Sign up today and get 1,000 free Crawlbase API requests — MCP stays free.
Most AI Agents Miss the Point
LLMs are powerful — but blind. Without real-time, contextual data, your AI agents are guessing instead of knowing.
Manual copy-pasting into prompts
Suggestions with no runtime awareness
Hallucinations from stale context
Agents that break when data changes
We have the solution!
Crawlbase Model Context Protocol (MCP) fixes all of that by plugging live web data straight into your agents.
It bridges the gap between web scraping and AI reasoning, so your LLMs are grounded in the NOW.
How It works: 1-minute setup
Select Your Tool
Claude Desktop · Cursor IDE · WindSurf IDE
Add the MCP Server
Copy your tool’s config from the panel and paste it into:
Claude Desktop:
Go to File → Settings → Developer → Edit Config → claude_desktop_config.json
Cursor IDE:
Go to File → Preferences → Cursor Settings → Tools and Integrations → New MCP Server → mcp.json
WindSurf IDE:
Go to File → Preferences → WindSurf Settings → General → MCP Servers → Manage MCPs → View Raw Config → mcp_config.json
{
"mcpServers": {
"crawlbase": {
"type": "stdio",
"command": "npx",
"args": [
"@crawlbase/mcp@latest"
],
"env": {
"CRAWLBASE_TOKEN": "[CRAWLBASE_TOKEN]",
"CRAWLBASE_JS_TOKEN": "[CRAWLBASE_JS_TOKEN]"
}
}
}
}
Add Your Crawlbase Token
Get your token from the Crawlbase Dashboard → Replace it with Crawlbase_Token in your config
🔑 Claim My 1,000 Free RequestsStart Crawling Like a Pro
crawl - Get live HTML
crawl_markdown - Extract clean markdown from a URL
crawl_screenshot - Take a screenshot of a webpage
Why Choose Crawlbase MCP?
Crawlbase MCP is powered by the Crawlbase Crawling API— the most affordable and robust Web Scraping API in the market.
Scales With You
Automate multi-step scraping and reasoning flows
Smarter Assistants
Give LLMs the data they need, when they need it
Free to Start
1,000 free Crawlbase API requests to start. MCP is always free.
No More Prompt Engineering Bloat
Skip the 20K-token dump
Trusted by customers worldwide
Data Operations Asia
“At Intel we need data at scale. Crawlbase has helped on our data demands allowing us to crawl billions of documents in a short period of time.”
Data Operations Asia
Chief Technical Officer
“We've been using Crawlbase for several years now to help us power parts of our aggregation pipeline with some key information that we wouldn't otherwise be able to get through traditional means.”
Chief Technical Officer
Chief Executive Officer
“Crawlbase helped us scale our data scraping needs in a fast, easy and cost effective way.”
Chief Executive Officer
Chief Executive Officer
“It’s critical to the success of our business to gather YouTube data regularly. Crawlbase has helped UpscaleMethod consistently meet demands for reviews and analytics. Thanks for providing such a solid service!”
Chief Executive Officer
Chief Executive Officer
“Crawlbase helps PageWatch to test sites that would otherwise be very difficult to crawl, and helps us be more confident in the results we pass on to our users.”
Chief Executive Officer
Chief Technical Officer
“Crawlbase makes collecting all of this data incredibly easy. It dramatically simplified the crawling and scraping process. Instead of handling proxy management, infrastructure and dozens of ever-changing re-captcha-systems ourselves, we delegate to the simple but powerful Crawlbase API and just get the problem solved.”
Chief Technical Officer
Trusted by more than 46,000 paying customers
Start crawling the web today
Try it free. No credit card required. Instant set-up.
