Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@ZeroCrawl MCP Search for the latest AI news and scrape the top 5 results"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Deploy your own private instance of ZeroCrawl in seconds:
Related MCP server: mcp-services
ZeroCrawl MCP is a 100% free forever web crawling and scraping server designed for the Model Context Protocol (MCP). It allows AI agents, LLM tools, and IDEs like Claude Desktop and Cursor to navigate the web with zero budget. It requires no paid API keys and operates seamlessly within the limits of Vercel's Free Tier.
Tool Name | Parameters | Description | Output Format |
|
| Fetches URL, cleans DOM with Readability, converts to Markdown, auto-fallback to Jina Reader on 403/SPA. | Markdown |
|
| Recursive same-domain link crawler with depth control and concurrency caps. | JSON |
|
| Structured OpenGraph, Twitter Cards, schema JSON-LD, and canonical data. | JSON |
|
| High-res full-page screenshot URL via Microlink + image URL extraction with alt tags. | JSON |
|
| XML sitemap and sitemap index parser. | JSON |
|
| Concurrent multi-URL scraping engine. | Array of Markdown |
|
| Fetch web search results and concurrently scrape each target link. | JSON |
We highly recommend utilizing our 1-Click Autonomous Agent Setup to completely eliminate manual configuration.
Don't want to configure this manually? Copy and paste the prompt below directly into your AI coding assistant (like Cursor, Windsurf, or Google Antigravity). Your AI will autonomously install, configure, and deploy the ZeroCrawl MCP server for you!
Copy & Paste this into your AI Agent:
Please install, configure, and deploy the ZeroCrawl MCP Server to my Vercel account. Execute these exact steps sequentially: 1. Run `git clone https://github.com/AgentNex/zerocrawl-mcp.git` and `cd zerocrawl-mcp`. 2. Run `npm install` to install all necessary packages. 3. Copy `.env.example` to `.env` and generate a secure random string for `AUTH_TOKEN`. 4. Run `npx vercel link --yes` to link the project to my Vercel account. 5. Deploy the project to production by running `npx vercel deploy --prod`. 6. Once successfully deployed, please configure my local MCP settings (in Claude Desktop or Cursor) to point to the new Vercel production URL at the `/api/mcp` endpoint using the `AUTH_TOKEN` you generated.
ZeroCrawl is built from the ground up to operate within the constraints of Vercel's Free Tier:
Sub-9s Timeout Safeguards: Operations automatically gracefully downgrade or return partial data to prevent serverless timeouts.
In-Memory TTL Caching Layer: Eliminates redundant fetching and scraping by returning instant responses for recent queries within a 10-minute window.
Zero-Chromium Architecture: Bypasses the need for heavy headless browsers. It leverages smart HTTP fetchers, DOM cleanup with Mozilla's Readability, and Turndown for clean Markdown generation. For complex SPAs or blocks, it automatically falls back to Jina Reader APIs.
MIT License (2026). See LICENSE for details.
This server cannot be deployed
Free web search for AI agents. No API key required. Hosted MCP in active development.
Web scraping for agents. Point it at a URL and it returns the page as clean markdown, JavaScript-rendered pages included. Point it at a site and it maps the URLs or crawls the section you need in the background, a few pages at a time so results fit in the conversation. Search the web and read full pages, extract fields with a JSON schema you define (validated, never invented), read a store's catalogue or a blog's posts from the platform's own feed, and check whether a page has changed. Failed requests cost nothing. The free plan includes 1,500 credits a month.
Zenrows MCP server — Fetch, Extract, Batch, and Browser Sessions for AI coding assistants
Web scraping for AI agents: scrape, search, crawl, map any website to markdown + JSON. No browser.
