One-click MCP servers with Cloudflare

One thing I’ve never liked about the whole MCP thing is the fact that you have to build a server and host it yourself. So ever since hearing about this, I’ve been looking for a self-contained solution where I could basically just describe the functionality that I want and the infrastructure could be handled itself.

It turns out Cloudflare actually has a solution for doing this, and I just love this about Cloudflare. I’ve actually talked about this elsewhere where I talk about how Cloudflare is basically eating the world by doing one-off services really well.

What are MCP servers?

Model Context Protocol (MCP) servers are a way to extend AI assistants with custom tools and data sources. They let you give your AI assistant access to specific capabilities—like querying databases, calling APIs, or performing specialized tasks. The problem is, traditionally you need to:

• Set up a server
• Handle authentication
• Manage scaling
• Deal with infrastructure
• Maintain uptime

This is a lot of overhead when you just want to add a simple capability to your AI workflow.

Enter Cloudflare’s one-click solution

Cloudflare Workers provides the perfect platform for MCP servers because:

• No infrastructure management – Cloudflare handles all the scaling and distribution
• Global edge network – Your MCP server runs close to users everywhere
• Simple deployment – Push code and it’s live
• Pay-per-use pricing – No paying for idle servers

Building a working MCP server

Let’s build an actual MCP server that I can use. I’ll create a simple “website analyzer” that can fetch and analyze any website’s content.

Step 1: Set up the project

mkdir cloudflare-mcp-analyzer
cd cloudflare-mcp-analyzer
npm init -y
npm install @modelcontextprotocol/sdk wrangler

Step 2: Create the MCP server

Create src/index.ts:

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
  CallToolRequestSchema,
  ListToolsRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

const server = new Server(
  {
    name: "website-analyzer",
    version: "1.0.0",
  },
  {
    capabilities: {
      tools: {},
    },
  },
);

// Define our tool
server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [
    {
      name: "analyze_website",
      description: "Analyze a website and extract key information",
      inputSchema: {
        type: "object",
        properties: {
          url: {
            type: "string",
            description: "The URL to analyze",
          },
        },
        required: ["url"],
      },
    },
  ],
}));

// Handle tool execution
server.setRequestHandler(CallToolRequestSchema, async (request) => {
  if (request.params.name === "analyze_website") {
    const { url } = request.params.arguments as { url: string };

    try {
      const response = await fetch(url);
      const html = await response.text();

      // Extract basic info
      const titleMatch = html.match(/.*?)<\/title>/i);
      const title = titleMatch ? titleMatch[1] : "No title found";

      const metaDescription = html.match(
        /\s+name="description"\s+content="(.*?)"/i,
      );
      const description = metaDescription
        ? metaDescription[1]
        : "No description found";

      // Count common elements
      const linkCount = (html.match(/\s/gi) || []).length;
      const imageCount = (html.match(/\s/gi) || []).length;
      const headingCount = (html.match(/[1-6]\s/gi) || []).length;

      return {
        content: [
          {
            type: "text",
            text: JSON.stringify(
              {
                url,
                title,
                description,
                stats: {
                  links: linkCount,
                  images: imageCount,
                  headings: headingCount,
                  contentLength: html.length,
                },
              },
              null,
              2,
            ),
          },
        ],
      };
    } catch (error) {
      return {
        content: [
          {
            type: "text",
            text: `Error analyzing website: ${error.message}`,
          },
        ],
      };
    }
  }

  throw new Error(`Unknown tool: ${request.params.name}`);
});

// For Cloudflare Workers
export default {
  async fetch(request: Request): Promise<Response> {
    if (request.method === "OPTIONS") {
      return new Response(null, {
        headers: {
          "Access-Control-Allow-Origin": "*",
          "Access-Control-Allow-Methods": "POST, OPTIONS",
          "Access-Control-Allow-Headers": "Content-Type",
        },
      });
    }

    // Handle MCP requests over HTTP
    const transport = new StdioServerTransport();
    await server.connect(transport);

    return new Response("MCP Server Running", { status: 200 });
  },
};

name = "mcp-website-analyzer"
main = "src/index.ts"
compatibility_date = "2024-01-01"

[build]
command = "npm run build"

[build.upload]
format = "modules"
main = "./index.js"

# Login to Cloudflare
wrangler login

# Deploy the worker
wrangler deploy

{
  "mcpServers": {
    "website-analyzer": {
      "url": "https://mcp-website-analyzer.YOUR-SUBDOMAIN.workers.dev",
      "description": "Analyzes websites and extracts key information"
    }
  }
}