mcpField
M
Model Context Protocol (MCP) Server + Strava OAuth

Model Context Protocol (MCP) Server + Strava OAuth

Local
github.com/kw510Developer Tools26Updated 80w ago

About Model Context Protocol (MCP) Server + Strava OAuth

An MCP server built on Cloudflare Workers with built-in Strava OAuth sign-in. Users connect to the server by logging in with their Strava account before accessing its tools. It acts as both an OAuth server for MCP clients and an OAuth client for Strava, and serves as a reference example for developers.

Key features

  1. 01Built-in Strava OAuth sign-in
  2. 02OAuth 2.1 server issuing and validating tokens
  3. 03Built on the workers-oauth-provider library
  4. 04Token storage in Cloudflare KV
  5. 05Remote MCP connections over an SSE endpoint

Use cases

  1. 01Providing Strava account sign-in for MCP clients
  2. 02Learning how to connect an OAuth provider to a remote MCP server
  3. 03Deploying an authenticated MCP server on Cloudflare Workers

MCP support

The project is a remote MCP server that exposes tools over an SSE endpoint to MCP clients such as Claude and Cursor, requiring Strava authorization before connection.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"strava-mcp-kw510": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/inspector"
]
}
}
}
Run commandnpx -y @modelcontextprotocol/inspector

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@modelcontextprotocol/inspector

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Model Context Protocol (MCP) Server + Strava OAuth

This is a Model Context Protocol (MCP) server that supports remote MCP connections, with Strava OAuth built-in. It allows users to connect to your MCP server by signing in with their Strava account.

Overview

The MCP server (powered by Cloudflare Workers) serves two roles:

  • Acts as an OAuth Server for your MCP clients
  • Acts as an OAuth Client for Strava's OAuth services

This project serves as a reference example for integrating OAuth providers with an MCP server deployed to Cloudflare, using the workers-oauth-provider library.

Prerequisites

  • A Strava account
  • A Cloudflare account
  • Node.js and npm installed
  • Wrangler CLI installed (npm install -g wrangler)

Quick Start

  1. Clone the repository:

    git clone https://github.com/kw510/strava-mcp.git
    cd strava-mcp
    npm install
    
  2. Set up your Strava API credentials (see Setting Up Strava API Credentials)

  3. Set up your Cloudflare KV namespace:

    wrangler kv:namespace create "OAUTH_KV"
    

    Update the wrangler.toml file with the generated KV ID.

  4. Deploy to Cloudflare:

    wrangler deploy
    

Setting Up Strava API Credentials

For Production
  1. Go to Strava's API Settings and create a new application

  2. Configure your application:

    • Application Name: Choose a name for your application
    • Category: Select an appropriate category
    • Website: Your website URL
    • Application Description: Brief description of your application
    • Authorization Callback Domain: mcp-strava-oauth.<your-subdomain>.workers.dev
    • Authorization Callback URL: https://mcp-strava-oauth.<your-subdomain>.workers.dev/callback
  3. Set your production environment variables:

    wrangler secret put STRAVA_CLIENT_ID
    wrangler secret put STRAVA_CLIENT_SECRET
    
For Development
  1. Create a separate Strava API application for development

  2. Configure your development application:

    • Authorization Callback Domain: localhost
    • Authorization Callback URL: http://localhost:8788/callback
  3. Create a .dev.vars file in your project root:

    STRAVA_CLIENT_ID=your_development_strava_client_id
    STRAVA_CLIENT_SECRET=your_development_strava_client_secret
    

Testing Your MCP Server

Using Inspector
  1. Install the Inspector tool:

    npx @modelcontextprotocol/inspector@latest
    
  2. Connect to your server:

    • For production: https://mcp-strava-oauth.<your-subdomain>.workers.dev/sse
    • For development: http://localhost:8788/sse
Using Claude Desktop
  1. Open Claude Desktop and go to Settings -> Developer -> Edit Config
  2. Add your MCP server configuration:
    {
      "mcpServers": {
        "strava": {
          "command": "npx",
          "args": [
            "mcp-remote",
            "https://mcp-strava-oauth.<your-subdomain>.workers.dev/sse"
          ]
        }
      }
    }
    
  3. Restart Claude Desktop and complete the OAuth flow

Development

Local Development
  1. Start the development server:

    wrangler dev
    
  2. The server will be available at http://localhost:8788

API Rate Limits

The Strava API has the following rate limits:

  • 200 requests every 15 minutes
  • 2,000 requests per day

How It Works

OAuth Provider

The OAuth Provider library handles:

  • OAuth 2.1 server implementation
  • Token issuance and validation
  • Secure token storage in KV
  • Strava OAuth integration
Durable MCP

Provides:

  • Persistent state management
  • Secure authentication context storage
  • User information access via this.props
  • Conditional tool availability
MCP Remote

Enables:

  • Client-server communication
  • Tool definition and management
  • Request/response serialization
  • SSE connection maintenance

Troubleshooting

  • If you see error messages in Claude Desktop, verify the connection by hovering over the 🔨 icon
  • For Cursor integration, use the "Command" type and combine command and args into one string
  • Ensure your callback URLs match exactly with what's configured in your Strava application

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.