MCP Server
Description
# MCP Server A minimal reference implementation of the Model Context Protocol (MCP) with a few mock tools and an optional Gradio playground. [](https://colab.research.google.com/github/zack-dev-cm/mcp_serve…
About
# MCP Server A minimal reference implementation of the Model Context Protocol (MCP) with a few mock tools and an optional Gradio playground. [](https://colab.research.google.com/github/zack-dev-cm/mcp_server/blob/main/MCP_colab.ipynb) ## Usage…
Details
- Author
- zack-dev-cm
- Downloads
- 247
- Categories
- Other
Jump to
- Reference implementation of the Model Context Protocol
- Mock tools for demonstration and testing
- Optional Gradio playground UI
- Google Colab notebook integration
- Plugin support for extending with new tools (e.g., OpenAI models)
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Run locally with python server.py. For Google Colab, use the helper function launch_in_colab() from the colab_adapter module after cloning the repo and installing dependencies. The API is served on port 8000 by default and the Gradio UI uses GRADIO_SERVER_PORT or the first free port starting at 7860.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp server": {
"mcp-server-zack-dev-cm": {
"command": "python",
"args": [
"server.py"
]
}
}
}
}McpServers
{
"mcp-server-zack-dev-cm": {
"command": "python",
"args": [
"server.py"
]
}
}MCP Server
A minimal reference implementation of the Model Context Protocol (MCP) with a few mock tools and an optional Gradio playground.
Usage
Run the server locally:
python server.py
Inside Google Colab or other notebooks use the helper:
from colab_adapter import launch_in_colab
launch_in_colab()
The API is served on port 8000 by default and the Gradio UI will try to use GRADIO_SERVER_PORT or the first free port starting at 7860.
Run in Google Colab
Open MCP_colab.ipynb in Colab or click the badge above and run the cells.
If starting from a blank notebook, run these commands to set up and launch the
server:
!git clone https://github.com/zack-dev-cm/mcp_server.git
%cd /content/mcp_server
!pip install fastapi uvicorn[standard] gradio==4.* pydantic python-dotenv
from colab_adapter import launch_in_colab
launch_in_colab()
You can now query the API from another cell:
import requests, time
time.sleep(2)
print(requests.get("http://localhost:8000/v1/resources").json())
The server output shows a public URL for the Gradio interface so you can try the demo visually.
LLM/VLM Plugin Example
Plugins can extend the server with new tools. The included openai_chat and openai_vision plugins show how to call OpenAI models. Set OPENAI_API_KEY in your environment and start the server:
pip install openai
python server.py
Invoke the openai.chat or openai.vision tools via the API or Gradio UI.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
