MCP Boilerplate — DevOps skill for Claude Code

MCP Boilerplate: Simple Setup Guide

This project helps you create your own remote MCP server on Cloudflare with user login and payment options. You don't need to be a technical expert to get it running.

[!NOTE] This project is now free to use and open source. If you want to support me, just follow me on X [@iannuttall](https://x.com/iannuttall) and subscribe to [my newsletter](https://ian.is).

What You'll Get

• An MCP server that works with Cursor, Claude and other AI assistants
• User login with Google or GitHub
• Payment processing with Stripe
• The ability to create both free and paid MCP tools

Setup Checklist

Before starting, make sure you have:

• Node.js installed (download from nodejs.org)
• A Cloudflare account (sign up at dash.cloudflare.com/sign-up)
• A Google account for setting up login (or GitHub if you prefer)
• A Stripe account for payments (sign up at dashboard.stripe.com/register)

Step-by-Step Setup

Step 1: Get the Code

1. Clone this repository to your computer:

git clone https://github.com/iannuttall/mcp-boilerplate.git
cd mcp-boilerplate

1. Install everything needed:

npm install

Step 2: Set Up the Database

1. Install Wrangler (Cloudflare's tool) if you haven't already:

npm install -g wrangler

1. Create a database for user login:

npx wrangler kv namespace create "OAUTH_KV"

Note: you can't use a different name for this database. It has to be "OAUTH_KV".

1. After running this command, you'll see some text that includes `id` and `preview_id` values

2. Open the `wrangler.jsonc` file in the project folder

3. Look for the section with `"kv_namespaces": [`

4. Add your database information there:

"kv_namespaces": [
  {
    "binding": "OAUTH_KV",
    "id": "paste-your-id-here",
    "preview_id": "paste-your-preview-id-here"
  }
]

Step 3: Set Up Your Local Settings

1. Create a file for your settings:

cp .dev.vars.example .dev.vars

1. Open the `.dev.vars` file in your code editor

2. You'll need to add several values here (we'll get them in the next steps)

Step 4a: Setting Up Google Login (Recommended)

1. Go to the Google Cloud Console
2. Create a new project with any name you like
3. Go to "APIs & Services" > "Credentials"
4. Click "+ CREATE CREDENTIALS" and choose "OAuth client ID"
5. If prompted, set up a consent screen:
   • Choose "External" for User Type
   • Add an App name (like "My AI Tool")
   • Add your email address where required
   • You can skip the "Scopes" and "Test users" sections
6. For the OAuth client:
   • Select "Web application" for Application type
   • Give it a name
   • Under "Authorized redirect URIs" add the following:

http://localhost:8787/callback/google

1. Click "CREATE"
2. You'll now see your Client ID and Client Secret - copy these val