Use This MCP server To

Demonstrate MCP server integration using Python SDK Run development inspector for MCP server debugging Install MCP server plugin in Claude LLM environment Enable MCP server discovery in GitHub Copilot Test MCP server functionality with real LLM hosts Troubleshoot MCP server installation and configuration

README

MCP Python Demo

This repository demonstrates the use of Model Context Protocol (MCP) SDK for Python.

Running Development Inspector

To run the development inspector:

mcp dev server.py

Installing in Claude

To install this MCP plugin in Claude:

Run this command:
```
mcp install server.py
```
Restart Claude
Troubleshooting: If you get an error, fix the uv path in the config to an absolute path:
```
which uv
```
Then update the configuration with the absolute path.

Installing in GitHub Copilot

📺 If you prefer learning from videos, watch this video from Burke Holland or this video from James Montemagno.

If you already installed in Claude Desktop

Enable chat.mcp.discovery.enabled: true in your settings and VS Code will discover existing MCP server lists, and proceed to use the tool in GitHub Copilot Agent mode.

If you did not install in Claude Desktop

If you want to associate the MCP server only with a particular repo, create a .vscode/mcp.json file with this content:

{
"inputs": [
],
"servers": {
      "pamelas-mcp": {
         "command": "PATH/TO/uv",
         "args": [
            "--directory",
            "/PATH/TO/mcp-python-demo",
            "run",
            "server.py"
         ]
      }
}
}

Alternatively, if you want to associate the MCP server with all repos, add to your VS Code User Settings JSON:

{
"mcp": {
   "inputs": [],
   "servers": {
      "pamelas-mcp": {
      "command": "/PATH/TO/uv",
      "args": [
         "--directory",
         "/PATH/TO/mcp-python-demo",
         "run",
         "server.py"
      ]
      }
   }
}
}

Another way to update settings is to run this command in the terminal:

code-insiders --add-mcp "{\"name\":\"pamelas-mcp\",\"command\":\"/PATH/TO/uv\",\"args\":[\"--directory\",\"/PATH/TO/mcp-python-demo\",\"run\",\"server.py\"]}"

Using tools in Copilot

Now that the mcp server is discoverable, open GitHub Copilot and select the Agent mode (not Chat or Edits).
Select the "refresh" button in the top right corner of the Copilot chat text field to refresh the server list.
Select the "🛠️" button to see all the possible tools, including the ones from this repo.
Put a question in the chat that would naturally invoke one of the tools, for example: "What is the current weather in Seattle?".

mcp-python-demo FAQ

How do I run the development inspector for mcp-python-demo?

Use the command 'mcp dev server.py' to start the development inspector.

How can I install the mcp-python-demo server in Claude?

Run 'mcp install server.py', then restart Claude to complete installation.

What should I do if I encounter a uv path error during installation?

Use 'which uv' to find the absolute path and update the configuration accordingly.

How do I enable MCP server discovery in GitHub Copilot?

Enable 'chat.mcp.discovery.enabled: true' in VS Code settings to discover MCP servers.

Are there video resources to learn about mcp-python-demo?

Yes, videos by Burke Holland and James Montemagno are available on YouTube for guidance.

Can mcp-python-demo be used for testing MCP server features?

Yes, it is designed for development, testing, and demonstration of MCP server capabilities.

Is mcp-python-demo compatible with multiple LLM hosts?

Yes, it supports integration with hosts like Claude and GitHub Copilot.