MCP GitHub OAuth Analytics Server
A Model Context Protocol server that enables AI models to access tools through GitHub authentication and provides comprehensive analytics tracking of tool usage, performance, and user behavior.
README
Model Context Protocol (MCP) Server + GitHub OAuth + Analytics
This is a Model Context Protocol (MCP) server that supports remote MCP connections, with GitHub OAuth authentication and built-in analytics tracking powered by MCP Analytics.
You can deploy it to your own Cloudflare account, and after you create your own GitHub OAuth client app, you'll have a fully functional remote MCP server with comprehensive analytics. Users will be able to connect to your MCP server by signing in with their GitHub account, and you'll get detailed insights into tool usage, performance, and user behavior.
Features
- ✅ GitHub OAuth Authentication - Secure user authentication via GitHub
- ✅ Remote MCP Protocol - Full MCP server implementation
- ✅ Analytics Tracking - Automatic tracking of tool usage, performance, and user metrics
- ✅ Access Control - Role-based tool access based on GitHub usernames
- ✅ Image Generation - AI-powered image generation for authorized users
- ✅ Production Ready - Deployed on Cloudflare Workers with Durable Objects
Analytics Dashboard
This server automatically tracks:
- 📊 Tool Usage - Which tools are used most frequently
- ⏱️ Performance Metrics - Execution times and success rates
- 👥 User Analytics - Active users and session data
- 🔧 Error Tracking - Failed requests and error details
- 💰 Revenue Tracking - Payment events (if using paid tools)
View your analytics at: https://mcpanalytics.dev
Getting Started
Clone the repo directly & install dependencies:
git clone <your-repo-url>
cd mcp-github-oauth-analytics
npm install
Setup Instructions
1. MCP Analytics Setup
- Sign up at https://mcpanalytics.dev
- Create a new project and get your API key
- Add the API key to your environment:
# For production
wrangler secret put MCP_ANALYTICS_API_KEY
# For local development (.dev.vars file)
MCP_ANALYTICS_API_KEY=your_api_key_here
2. GitHub OAuth Setup
For Production
Create a new GitHub OAuth App:
- Homepage URL:
https://your-worker-name.your-subdomain.workers.dev - Authorization callback URL:
https://your-worker-name.your-subdomain.workers.dev/callback - Note your Client ID and generate a Client secret
Set secrets via Wrangler:
wrangler secret put GITHUB_CLIENT_ID
wrangler secret put GITHUB_CLIENT_SECRET
wrangler secret put COOKIE_ENCRYPTION_KEY # Use: openssl rand -hex 32
For Local Development
Create another GitHub OAuth App for development:
- Homepage URL:
http://localhost:8788 - Authorization callback URL:
http://localhost:8788/callback
Create a .dev.vars file:
GITHUB_CLIENT_ID=your_development_github_client_id
GITHUB_CLIENT_SECRET=your_development_github_client_secret
COOKIE_ENCRYPTION_KEY=your_random_encryption_key
MCP_ANALYTICS_API_KEY=your_analytics_api_key
3. KV Namespace Setup
# Create the KV namespace
wrangler kv:namespace create "OAUTH_KV"
# Update wrangler.toml with the returned KV ID
4. Configure Access Control
Edit the ALLOWED_USERNAMES in your main file to control who can access the image generation tool:
const ALLOWED_USERNAMES = new Set<string>([
'yourusername',
'teammate1',
'teammate2'
]);
Deployment
Deploy to Cloudflare Workers:
wrangler deploy
Your MCP server will be available at:
https://your-worker-name.your-subdomain.workers.dev/sse
Testing Your Server
Using MCP Inspector
npx @modelcontextprotocol/inspector@latest
Enter your server URL and test the authentication flow.
Using Claude Desktop
- Open Claude Desktop → Settings → Developer → Edit Config
- Add this configuration:
{
"mcpServers": {
"github-mcp": {
"command": "npx",
"args": [
"mcp-remote",
"https://your-worker-name.your-subdomain.workers.dev/sse"
]
}
}
}
- Restart Claude Desktop and complete the OAuth flow
- Test with: "Could you use the math tool to add 23 and 19?"
Using Other MCP Clients
Cursor: Use command format: npx mcp-remote https://your-worker-url/sse
Windsurf: Add the same JSON configuration as Claude Desktop
Available Tools
add (All Users)
Simple math addition tool for testing connectivity.
Usage: "Add 5 and 3"
generateImage (Authorized Users Only)
AI-powered image generation using Cloudflare's Flux model.
Usage: "Generate an image of a sunset over mountains"
Parameters:
prompt: Description of the image to generatesteps: Quality steps (4-8, higher = better quality)
Analytics Integration
This server uses the AnalyticsMcpAgent which automatically tracks:
- ✅ Tool Execution Times - How long each tool takes to run
- ✅ Success/Failure Rates - Which tools work reliably
- ✅ User Sessions - Who's using your server and when
- ✅ Parameter Logging - What inputs users are providing (safely sanitized)
- ✅ Error Details - Full error context for debugging
- ✅ GitHub User Data - Email and username from OAuth (for user analytics)
Viewing Analytics
- Visit https://dashboard.mcpanalytics.dev
- Sign in with the same account used to create your API key
- View real-time dashboards showing:
- Tool usage trends
- Performance metrics
- User activity
- Error rates and details
Local Development
Start the development server:
wrangler dev
Your server will be available at http://localhost:8788/sse
Architecture
OAuth Provider
The OAuth Provider library serves as a complete OAuth 2.1 server implementation, handling:
- MCP client authentication
- GitHub OAuth integration
- Token management and validation
- Secure state storage in Cloudflare KV
Analytics Agent
The AnalyticsMcpAgent extends the base MCP functionality with:
- Automatic event tracking for all tool calls
- User identification via OAuth props
- Performance monitoring and error tracking
- Safe parameter and result logging
Durable Objects
Provides persistent state management with:
- User session continuity
- Authentication context preservation
- Scalable real-time connections
Security & Privacy
- 🔒 OAuth 2.1 - Industry standard authentication
- 🔐 Encrypted Tokens - All auth data encrypted in transit and storage
- 🛡️ Access Control - Fine-grained permissions per GitHub user
- 🧹 Data Sanitization - Sensitive data automatically redacted from logs
- ⏰ Token Expiration - Automatic token refresh and expiration
Troubleshooting
Common Issues
"Invalid API key" in analytics: Verify your MCP_ANALYTICS_API_KEY is set correctly
OAuth callback errors: Ensure your GitHub OAuth app URLs match your deployment URLs exactly
Tools not appearing: Check that the user's GitHub username is in ALLOWED_USERNAMES for restricted tools
Connection timeouts: Verify your Worker is deployed and responding at the correct URL
Support
- 📖 MCP Analytics Docs: https://docs.mcpanalytics.dev
- 💬 GitHub Issues: For bugs and feature requests
- 📧 Email Support: Available for Pro plan users
License
MIT License - see LICENSE file for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。