Crawl MCP Server
Connect any MCP-compatible AI agent — Cursor, Claude Desktop, Windsurf, Cline, and more — to the Crawl API. Your agent can scrape web pages, extract structured data, deep-crawl domains, run web searches, and check your credit balance in real time.
Install in 3 steps
Install the MCP server from npm
npm install -g @crawl/mcp
npm package will be available once the first publish completes. If it isn't found yet, use the source build below.
Get your API key
Copy an API key from your Dashboard. The key is only shown once.
Get your API keyAdd this server to your MCP client config
Cursor → Settings → MCP → Add new MCP server
Claude Desktop → Settings → Developer → Edit Config
Build from source
Install from the repository directly
git clone https://github.com/winbayai/Crawl.git cd Crawl/mcp-server npm install npm run build npm link
Publish to npm
To enable npm install, create an npm organization @crawl, generate an automation access token, and add it as the NPM_TOKEN secret in GitHub repo settings. CI will publish on the next push to main.
Example client config
Cursor
Cursor → Settings → MCP → Add new MCP server
{
"mcpServers": {
"crawl": {
"command": "crawl-mcp",
"args": [],
"env": {
"CRAWL_API_KEY": "sk_live_...",
"CRAWL_API_BASE": "https://api-crawl.top1.ai"
}
}
}
}Claude Desktop
Claude Desktop → Settings → Developer → Edit Config
{
"mcpServers": {
"crawl": {
"command": "crawl-mcp",
"env": {
"CRAWL_API_KEY": "sk_live_...",
"CRAWL_API_BASE": "https://api-crawl.top1.ai"
}
}
}
}Environment variables
CRAWL_API_KEY=sk_live_... CRAWL_API_BASE=https://api-crawl.top1.ai
Built-in tools
scrape
1 creditScrape a single URL to clean Markdown.
extract
3 creditsExtract structured data via LLM with a JSON schema.
crawl
1 / pageBFS deep-crawl within the same domain, billed by pages fetched.
search
2 creditsWeb search plus optional per-result scraping.
usage
freeCheck remaining credits and recent usage summary.
Ready to let your agent browse the web?
Sign up and get 100 free credits plus a default API key.