Skip to content

QuantGeekDev/docker-mcp

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🐳 docker-mcp

Python 3.12 License: MIT Code style: black

A powerful Model Context Protocol (MCP) server for Docker operations, enabling seamless container and compose stack management through Claude AI.

✨ Features

  • 🚀 Container creation and instantiation
  • 📦 Docker Compose stack deployment
  • 🔍 Container logs retrieval
  • 📊 Container listing and status monitoring

🎬 Demos

Deploying a Docker Compose Stack

docker-mcp-demo-1.mp4

Analyzing Container Logs

docker-mcp-demo-2.mp4

🚀 Quickstart

To try this in Claude Desktop app, add this to your claude config files:

{
  "mcpServers": {
    "docker-mcp": {
      "command": "uvx",
      "args": [
        "docker-mcp"
      ]
    }
  }
}

Prerequisites

  • UV (package manager)
  • Python 3.12+
  • Docker Desktop or Docker Engine
  • Claude Desktop

Installation

Claude Desktop Configuration

Add the server configuration to your Claude Desktop config file:

MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

💻 Development Configuration
{
  "mcpServers": {
    "docker-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "<path-to-docker-mcp>",
        "run",
        "docker-mcp"
      ]
    }
  }
}
🚀 Production Configuration
{
  "mcpServers": {
    "docker-mcp": {
      "command": "uvx",
      "args": [
        "docker-mcp"
      ]
    }
  }
}

🛠️ Development

Local Setup

  1. Clone the repository:
git clone https://github.com/QuantGeekDev/docker-mcp.git
cd docker-mcp
  1. Create and activate a virtual environment:
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
  1. Install dependencies:
uv sync

🔍 Debugging

Launch the MCP Inspector for debugging:

npx @modelcontextprotocol/inspector uv --directory <path-to-docker-mcp> run docker-mcp

The Inspector will provide a URL to access the debugging interface.

📝 Available Tools

The server provides the following tools:

create-container

Creates a standalone Docker container

{
    "image": "image-name",
    "name": "container-name",
    "ports": {"80": "80"},
    "environment": {"ENV_VAR": "value"}
}

deploy-compose

Deploys a Docker Compose stack

{
    "project_name": "example-stack",
    "compose_yaml": "version: '3.8'\nservices:\n  service1:\n    image: image1:latest\n    ports:\n      - '8080:80'"
}

get-logs

Retrieves logs from a specific container

{
    "container_name": "my-container"
}

list-containers

Lists all Docker containers

{}

🚧 Current Limitations

  • No built-in environment variable support for containers
  • No volume management
  • No network management
  • No container health checks
  • No container restart policies
  • No container resource limits

🤝 Contributing

  1. Fork the repository from docker-mcp
  2. Create your feature branch
  3. Commit your changes
  4. Push to the branch
  5. Open a Pull Request

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

✨ Authors

  • Alex Andru - Initial work | Core contributor - @QuantGeekDev
  • Ali Sadykov - Initial work | Core contributor - @md-archive

Made with ❤️

About

A docker MCP Server (modelcontextprotocol)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages