MCP Streamable HTTP Server

What is MCP?

Model Context Protocol (MCP) is an open standard that enables seamless integration between AI assistants and external data sources, tools, and services. MCP acts as a universal bridge, allowing AI models to interact with various tools and databases in a standardized way.

Related MCP server: G4 Data Model MCP Server

Why We Created This

This project demonstrates how to deploy MCP servers over HTTP, making powerful tools accessible to AI assistants like GitHub Copilot, Claude, and other AI systems. By deploying MCP servers remotely, we can:

• Scale globally: Deploy once, use everywhere

• Share tools: Multiple users can access the same tool endpoints

• Simplify setup: No local installation required

• Enhance AI capabilities: Extend AI assistants with custom functionality

🚀 Live Deployed Servers

We have deployed 3 MCP servers that are ready to use! Each server provides different functionality:

🔧 Available Servers

📝 Server Details

Echo Server

• Purpose: Text processing and echo functionality

• Tools: Echo messages, text transformation, string manipulation

• Use Cases: Testing MCP connections, simple text operations

Math Server

• Purpose: Mathematical computations and calculations

• Tools: Basic arithmetic, advanced math functions, statistical operations

• Use Cases: Calculations, data analysis, mathematical problem solving

Social Server

• Purpose: Social media and content tools

• Tools: Content generation, social media formatting, text enhancement

• Use Cases: Social media posts, content creation, text optimization

🔌 Connect to GitHub Copilot

To use these servers with GitHub Copilot, follow these steps:

1. Open your MCP configuration file (mcp.json) for GitHub Copilot

2. Add the following server configurations:

{
  "mcpServers": {
    "maths_tools": {
      "url": "https://stream-mcp.onrender.com/math/mcp/",
      "type": "http"
    },
    "social_tool": {
      "url": "https://stream-mcp.onrender.com/Social/mcp/",
      "type": "http"
    },
    "echo": {
      "url": "https://stream-mcp.onrender.com/echo/mcp/",
      "type": "http"
    }
  }
}

3. Save the configuration and restart GitHub Copilot

4. Start using the tools in your AI conversations!

🛠️ Local Development

If you want to run the servers locally or contribute to the project:

Prerequisites

• Python 3.8+

• pip or uv package manager

Quick Start

1. Clone the repository:

   git clone https://github.com/ai-engineer-devansh-singh/stream_mcp.git
   cd stream_mcp

2. Install dependencies:

   pip install -r requirements.txt

3. Run the combined server:

   python Example/server.py

4. Access local endpoints:

   • Echo: http://localhost:10000/echo/mcp/

   • Math: http://localhost:10000/math/mcp/

   • Social: http://localhost:10000/Social/mcp/

📁 Project Structure

├── Example/
│   ├── echo_server.py      # Echo MCP server implementation
│   ├── math_server.py      # Math MCP server implementation
│   ├── Social.py           # Social MCP server implementation
│   └── server.py           # Combined FastAPI server
├── docs/                   # Documentation and assets
├── pyproject.toml          # Project configuration
├── runtime.txt             # Python runtime specification
└── server.py               # Basic MCP server example

💡 Need Help?

If you encounter any issues while setting up or using these MCP servers, please don't hesitate to reach out:

• GitHub Issues: Create an issue in this repository

• Contact: Feel free to contact me directly for any problems or questions

• LinkedIn: Follow me on LinkedIn for updates and AI engineering content

🤝 Contributing

Contributions are welcome! Feel free to:

• Report bugs

• Suggest new features

• Submit pull requests

• Improve documentation

📜 License

This project is open source and available under the MIT License.