MCP Startup Framework
A comprehensive framework for building remote MCP servers on Cloudflare Workers featuring OAuth 2.1 authentication, PostgreSQL integration, and Stripe-powered paid tools. It enables developers to create scalable, platform-agnostic MCP tools with support for user registration and subscription-based functionality.
README
MCP Startup Framework
A complete framework for building remote Model Context Protocol (MCP) servers on Cloudflare Workers with OAuth authentication, PostgreSQL database, and Stripe-powered paid tools.
🚀 Features
- 🔐 OAuth 2.1 Provider - Complete authentication with user registration/login
- 🗄️ PostgreSQL Integration - Scalable production-ready database with connection pooling
- 💳 Paid Tools Framework - Stripe-powered premium tools with subscription support
- 🌐 Platform Agnostic - Optimized for Cloudflare Workers but deployable anywhere (Vercel, Railway, AWS, self-hosted)
- 🛠️ MCP Tools Framework - Modular tool system with user context
- 🔌 REST API Routes - Easy-to-use system for adding custom endpoints
- 🎨 Custom Views - Extensible UI system for privacy policies, ToS, documentation pages
- 📡 Streamable HTTP Transport - Works with
mcp-remotefor local Claude Desktop integration - 📱 Cross-Platform - Works on desktop, web, and mobile MCP clients
⚡ Quick Start
Prerequisites
- Node.js 18+
- Cloudflare account
- PostgreSQL database (see database setup)
- Stripe account (for paid tools)
Installation
# Clone the repository
git clone https://github.com/f/mcp-startup-framework
cd mcp-startup-framework
# Install dependencies
npm install
# Copy environment template
cp .dev.vars.example .dev.vars
Configuration
Edit .dev.vars with your settings:
DATABASE_URL="postgresql://username:password@host:port/database"
JWT_SECRET="your-super-secret-jwt-key"
COOKIE_ENCRYPTION_KEY="32-character-encryption-key"
STRIPE_SECRET_KEY="sk_test_your_stripe_key"
STRIPE_PRICE_ID_FOR_PREMIUM_MATH="price_your_stripe_price_id"
BASE_URL="https://your-domain.workers.dev"
Development
# Start development server
npm run dev
# Initialize database (in another terminal)
curl -X POST http://localhost:8787/init-db
# Visit http://localhost:8787 to register/login
🛠️ Adding Tools & Routes
Everything starts from src/index.ts:
// Register MCP tools
backend
.registerTool(registerAddTool)
.registerTool(registerGreetingTool)
.registerTool(registerPremiumMathTool);
// Add REST API routes
backend
.route('GET', '/api/status', (c) => c.json({ status: 'ok' }))
.authRoute('GET', '/api/profile', (c, userContext) =>
c.json({ user: userContext })
);
See Tools Guide and Routes Guide for details.
Custom Views
Extend beyond MCP with custom pages:
// Add privacy policy, terms of service, documentation
backend
.route('GET', '/privacy', (c) => c.html(generatePrivacyPage()))
.route('GET', '/terms', (c) => c.html(generateTermsPage()))
.route('GET', '/docs', (c) => c.html(generateDocsPage()));
See Views Guide for creating custom pages.
🚀 Deployment
# Set production secrets
wrangler secret put JWT_SECRET
wrangler secret put COOKIE_ENCRYPTION_KEY
wrangler secret put STRIPE_SECRET_KEY
# Set custom price IDs per paid tool
wrangler secret put STRIPE_PRICE_ID_FOR_PREMIUM_MATH
# Deploy to Cloudflare
npm run deploy
See Deployment Guide for full instructions.
🔗 Integration Options
Using claude.ai (Pro Plan)
For claude.ai Pro users, you can integrate directly:
- Navigate to claude.ai settings
- Click "Add Integration"
- Enter your server URL:
https://your-domain.workers.dev/sse - Complete the OAuth flow
Using Claude Desktop
For local Claude Desktop app:
1. Install MCP Remote Client
npm install -g mcp-remote
2. Configure Claude Desktop
Open Claude Desktop settings: Settings > Developer > Edit Config
Add to your claude_desktop_config.json:
{
"mcpServers": {
"my-startup": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://your-domain.workers.dev/mcp"
]
}
}
}
3. Restart Claude Desktop
After saving, restart Claude Desktop. Your MCP tools will appear in the chat interface.
Transport Options
The framework supports multiple transport protocols:
- Streamable HTTP (default) -
/mcpendpoint - Server-Sent Events (SSE) -
/sseendpoint
Both work seamlessly with mcp-remote for local development and production use.
🌐 Platform Flexibility
While optimized for Cloudflare Workers, this framework runs anywhere:
- Vercel: Deploy with
@vercel/noderuntime - Railway: One-click deploy with built-in PostgreSQL
- AWS Lambda: Use with RDS or Aurora Serverless
- Google Cloud Run: Deploy with Cloud SQL
- Self-hosted: Run with Docker on any VPS
Simply update your database connection and deployment configuration!
Production Database
The framework works with any PostgreSQL provider:
- Neon - Serverless PostgreSQL with autoscaling
- Supabase - PostgreSQL with real-time features
- Railway - Managed PostgreSQL with automatic backups
- Amazon RDS - Enterprise-grade PostgreSQL
- Cloudflare Hyperdrive - Connection pooling for optimal performance
All providers support the scalability needed for production MCP servers.
📚 Documentation
- Setup Guide - Detailed installation and configuration
- Database Setup - PostgreSQL options and optimization
- Development Guide - Project structure and workflows
- Adding Tools - Create free and paid MCP tools
- Custom Routes - Add REST API endpoints
- Views Customization - Customize all UI pages (login, consent, payment, etc.)
- Deployment - Production deployment instructions
🤝 Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
📄 License
MIT License - see LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.