mcp-server-toolkit

jtabeling/mcp-server-toolkit

3.1

If you are the rightful owner of mcp-server-toolkit and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The Docker MCP Toolkit Exploration project sets up a Docker MCP server to enable AI-assisted Docker operations using the Model Context Protocol.

Tools
6
Resources
0
Prompts
0

Docker MCP Toolkit Exploration

This project demonstrates how to set up and use a Docker MCP (Model Context Protocol) server to enable AI-assisted Docker operations in Cursor.

What is MCP?

MCP (Model Context Protocol) is a protocol that allows AI assistants to connect to external data sources, tools, and services. This enables AI to perform real-time operations and access contextual information.

Features

This Docker MCP server provides the following tools:

Container Management

  • List Containers: View all Docker containers (running and stopped)
  • Start Container: Start a stopped container by name or ID
  • Stop Container: Stop a running container by name or ID
  • Remove Container: Remove a container (with optional force flag)

Image Management

  • List Images: View all Docker images on the system

System Information

  • Docker Info: Get detailed Docker system information

Setup Instructions

Prerequisites

  • Node.js (version 14 or higher)
  • Docker Desktop or Docker Engine
  • Cursor IDE

Installation

  1. Clone or create this project

    # If starting fresh
mkdir mcp-docker-exploration
cd mcp-docker-exploration

  2. Install dependencies

    npm install

  3. Test the MCP server

    node docker-mcp-server.js

Cursor Integration

To integrate with Cursor, you need to configure the MCP server in your Cursor settings:

  1. Open Cursor Settings

    • Go to Settings (Ctrl/Cmd + ,)
    • Search for "MCP" or "Model Context Protocol"

  2. Add MCP Server Configuration

    • Add the path to your docker-mcp-server.js file
    • Or use the provided cursor-mcp-config.json as a reference

  3. Restart Cursor

    • Restart Cursor to load the new MCP configuration

Usage Examples

Once configured, you can ask the AI assistant to:

  • "Show me all Docker containers"
  • "Start the container named 'my-app'"
  • "Stop the container with ID 'abc123'"
  • "Remove the stopped container 'old-container'"
  • "List all Docker images"
  • "Get Docker system information"

Available Tools

list_containers

Lists all Docker containers with options to show stopped containers.

Parameters:

  • all (boolean, optional): Show all containers including stopped ones (default: true)

list_images

Lists all Docker images on the system.

start_container

Starts a stopped Docker container.

Parameters:

  • container (string, required): Container name or ID to start

stop_container

Stops a running Docker container.

Parameters:

  • container (string, required): Container name or ID to stop

remove_container

Removes a Docker container.

Parameters:

  • container (string, required): Container name or ID to remove
  • force (boolean, optional): Force removal of running container (default: false)

docker_info

Gets detailed Docker system information.

Development

Adding New Tools

To add new Docker operations, modify the docker-mcp-server.js file:

  1. Add a new case in the tools/call handler
  2. Implement the corresponding method
  3. Add the tool definition to the tools/list handler

Testing

Test the server manually:

node docker-mcp-server.js

The server communicates via stdio, so you can test it by sending JSON messages to stdin.

Troubleshooting

Common Issues

  1. Docker not found

    • Ensure Docker is installed and running
    • Check that docker command is available in PATH

  2. Node.js not found

  3. MCP server not connecting

    • Check Cursor MCP configuration
    • Ensure the server file path is correct
    • Restart Cursor after configuration changes

Debug Mode

To run the server in debug mode, add logging:

console.error("Debug: Processing request:", request);

Security Considerations

  • The MCP server has access to Docker commands
  • Ensure proper permissions and access controls
  • Consider running in a restricted environment for production use

Next Steps

Potential enhancements:

  • Add Docker Compose support
  • Implement image building and pushing
  • Add network and volume management
  • Create a web interface for management
  • Add support for Docker Swarm

Resources