docker-mcp

To use docker-mcp with the Claude Desktop app, add the server configuration to your Claude Desktop configuration file. This allows Claude to perform Docker operations like container management and log retrieval.

Prerequisites:

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

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

Depending on your environment, add the configuration to the appropriate file path:

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

Production Configuration

Use this for standard usage with uvx:

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

Development Configuration

Use this when working on the local repository to run the server from a specific directory:

{
  "mcpServers": {
    "docker-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "<path-to-docker-mcp>",
        "run",
        "docker-mcp"
      ]
    }
  }
}

To develop on docker-mcp, follow these steps to set up your local environment:

1. Clone the repository:

   git clone https://github.com/QuantGeekDev/docker-mcp.git
   cd docker-mcp

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies using uv:

   uv sync

You can automatically install docker-mcp for Claude Desktop using the Smithery CLI:

npx @smithery/cli install docker-mcp --client claude

To debug the server, launch the MCP Inspector. This will provide a URL to access a web-based debugging interface.

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

The docker-mcp server exposes the following tools for managing Docker resources:

• create-container: Creates a standalone Docker container.
• deploy-compose: Deploys a Docker Compose stack.
• get-logs: Retrieves logs from a specific container.
• list-containers: Lists all Docker containers.

### create-container
```json
{
    "image": "image-name",
    "name": "container-name",
    "ports": {"80": "80"},
    "environment": {"ENV_VAR": "value"}
}

deploy-compose

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

get-logs

{
    "container_name": "my-container"
}

list-containers

{}

The following features are currently not supported by docker-mcp:

• Built-in environment variable support for containers
• Volume management
• Network management
• Container health checks
• Container restart policies
• Container resource limits

The main() function serves as the primary entry point for the docker-mcp package. It initializes and runs the MCP server using an asynchronous event loop. This is the function you should call if you are running the package as a script or integrating its execution into a Python application.

from docker_mcp import main

if __name__ == "__main__":
    main()