DockerHub MCP Server

A production-ready Model Context Protocol (MCP) server that provides comprehensive DockerHub integration for AI assistants. This server enables AI assistants to search, analyze, and manage Docker images through standardized MCP tools.

🚀 Features

Core Docker Hub Tools

• docker_search_images - Search Docker Hub for images

• docker_get_image_details - Get detailed information about an image

• docker_list_tags - List all tags for a repository

• docker_get_manifest - Retrieve image manifest

• docker_analyze_layers - Analyze image layers and sizes

• docker_compare_images - Compare two images (layers, sizes, base images)

• docker_get_dockerfile - Attempt to retrieve Dockerfile (when available)

• docker_get_stats - Get download statistics and star count

Advanced Tools

• docker_get_vulnerabilities - Fetch security scan results

• docker_get_image_history - Get image build history

• docker_track_base_updates - Check if base images have updates

• docker_estimate_pull_size - Calculate download size for an image

• docker_delete_tag - Delete a tag from your private repository

• docker_cache_info - Monitor cache performance and statistics

Key Capabilities

• ✅ MCP Client Compatible - Works with Claude Desktop, Cursor, Cline, and other MCP clients

• ✅ Private Registry Support - Access private DockerHub repositories with JWT authentication

• ✅ Smart Caching System - Intelligent TTL-based caching with LRU eviction for optimal performance

• ✅ Rate Limit Handling - Intelligent request management for DockerHub's API limits

• ✅ Comprehensive Error Handling - Graceful fallbacks and clear error messages

• ✅ TypeScript Implementation - Full type safety and excellent developer experience

• ✅ Extensive Testing - 26 tests covering all functionality and edge cases

Related MCP server: Docker Build MCP Server

🛠️ Installation & Setup

Prerequisites

• Node.js 18+

• npm or yarn

• DockerHub account (for private repositories)

Quick Start

1. Clone the repository:

   https://github.com/keshavjha123/DockerhubMCP.git
   cd DockerhubMCP

2. Install dependencies:

   npm install

3. Configure environment variables:

   cp .env.example .env
   # Edit .env with your DockerHub credentials

4. Build the project:

   npm run build

5. Start the server:

   npm start

⚙️ Configuration

Environment Variables

Create a .env file in the project root:

# DockerHub Authentication (for private repositories)
DOCKERHUB_USERNAME=your-username
DOCKERHUB_PASSWORD=your-password (same as personal access token)
DOCKERHUB_TOKEN=your-personal-access-token

# Optional: Custom API endpoints
DOCKERHUB_API_BASE=https://hub.docker.com/v2
DOCKER_REGISTRY_BASE=https://registry-1.docker.io

Authentication Methods

1. Username/Password: Traditional login credentials

2. Personal Access Token: Recommended for better security

3. No Authentication: For public repositories only

🐳 Docker Deployment

Quick Start with Docker

# One-command setup (Linux/macOS)
./docker-setup.sh

# One-command setup (Windows PowerShell)
.\docker-setup.ps1

Manual Docker Setup

Prerequisites: Make sure you're logged into Docker Hub to avoid rate limiting:

docker login

For Linux/macOS:

# 1. Configure environment (only if .env doesn't exist)
cp .env.example .env
# Edit .env with your DockerHub credentials

# 2. Build and run
docker-compose build
docker-compose up -d

# 3. Verify running
docker-compose ps

For Windows PowerShell:

# 1. Configure environment (only if .env doesn't exist)
Copy-Item .env.example .env
# Edit .env with your DockerHub credentials

# 2. Build and run
docker-compose build
docker-compose up -d

# 3. Verify running
docker-compose ps

⚠️ Note: Only run the first step if you don't already have a .env file with your credentials. The copy command will overwrite existing files.

✅ Docker Mode - TESTED & VERIFIED

The Docker deployment has been fully tested with:

• ✅ All 16 MCP tools functioning

• ✅ Private repository access working

• ✅ Smart caching and rate limiting active

• ✅ 26/26 tests passing

For detailed Docker deployment instructions, see Docker Deployment Guide.

🔌 MCP Client Integration

Claude Desktop

Standard Installation:

{
  "mcpServers": {
    "dockerhub": {
      "command": "node",
      "args": ["/path/to/pixil-mcp-server/dist/index.js"],
      "env": {
        "DOCKERHUB_USERNAME": "your-user-name",
        "DOCKERHUB_PASSWORD": "your-token-here",
        "DOCKERHUB_TOKEN": "your-token-here"
      }
    }
  }
}

To access a private repository, specify the full repository name in your Claude Desktop in your conversation (e.g., keshavmadhav12/keshavprivaterepo).

Alternatively, you can simply mention your DockerHub username to access your private repositories. Use the docker_list_repositories tool to retrieve the exact names of your private repositories.

Cursor/Cline

Configure the MCP server in your IDE settings to enable Docker image analysis capabilities.

📚 Usage Examples

Search for Images

Find the most popular Python images

Analyze Image Layers

Analyze the layers of nginx:latest and show optimization opportunities

Compare Images

Compare ubuntu:20.04 and ubuntu:22.04 and show the differences

Security Analysis

Check if my-org/my-app:latest has any critical vulnerabilities

Private Repository Management

List all tags for my-private-repo and delete outdated ones

🧪 Development & Testing

Development Mode

npm run dev

Run Tests

npm test

Linting

ESLint is configured to handle JavaScript files. TypeScript files are checked by the TypeScript compiler during npm run build.

To run linting:

npm run build  # TypeScript checking




### Test with Private Repositories
```bash
# Configure your credentials in .env first
node local_test/mcp-bulk-client-private.mjs

📖 API Documentation

Tool Schemas

All tools use Zod schemas for validation:

// Example: docker_search_images
inputSchema: z.object({
  query: z.string(),
  limit: z.number().optional()
})

outputSchema: z.object({
  images: z.array(z.object({
    name: z.string(),
    description: z.string(),
    stars: z.number(),
    pulls: z.number()
  }))
})

Error Handling

All errors are returned in MCP-compliant format:

{
  "error": {
    "code": "DOCKER_API_ERROR",
    "message": "Failed to fetch image details: 404 Not Found"
  }
}

🔒 Security Considerations

• Never commit credentials to version control

• Use personal access tokens instead of passwords

• Implement proper secret management in production

• Monitor API usage to prevent rate limit abuse

• Validate all inputs using Zod schemas

🚀 Production Deployment

Environment Setup

# Production environment variables
NODE_ENV=production
DOCKERHUB_TOKEN=production-token
LOG_LEVEL=info

Process Management

# Using PM2
pm2 start dist/index.js --name dockerhub-mcp

# Using systemd
sudo systemctl enable dockerhub-mcp
sudo systemctl start dockerhub-mcp

Monitoring

• Monitor API rate limits

• Track response times

• Log authentication failures

• Alert on service disruptions

🤝 Contributing

1. Fork the repository

2. Create a feature branch

3. Make your changes

4. Add tests for new functionality

5. Run the test suite

6. Submit a pull request

📋 Roadmap

Immediate Priorities

• Persistent caching with Redis (in-memory caching implemented)

• Rate limit visualization

• Batch operations

• Enhanced error recovery

Future Enhancements

• StreamableHTTP transport

• Multi-registry support

• Web dashboard

• Prometheus metrics

• Vulnerability severity filtering

• License compliance checking

� Troubleshooting

Common Issues

Authentication Failures:

• Verify your DockerHub credentials

• Check token permissions and scopes

• Ensure 2FA is properly configured

Rate Limiting:

• Use authenticated requests when possible

• Built-in smart caching reduces API calls automatically

• Intelligent request queuing handles high-volume usage

Connection Issues:

• Verify network connectivity to DockerHub

• Check firewall and proxy settings

• Validate API endpoint URLs

Debug Mode

DEBUG=* npm start

📄 License

MIT License - see LICENSE file for details.

🔗 Links

• MCP Specification

• TypeScript MCP SDK

• DockerHub API Documentation

• Docker Registry API

💬 Support

• Create an issue for bug reports

• Start a discussion for feature requests

• Check the documentation for common questions

Made with ❤️ for the Docker and AI community

🧰 MCP Tools Implemented

This server implements the following MCP tools:

• docker_search_images: Search Docker Hub for images

• docker_get_image_details: Get detailed information about an image

• docker_list_tags: List all tags for a repository

• docker_get_manifest: Retrieve image manifest

• docker_analyze_layers: Analyze image layers and sizes

• docker_compare_images: Compare two images (layers, sizes, base images)

• docker_get_dockerfile: Attempt to retrieve Dockerfile (when available)

• docker_get_stats: Get download statistics and star count

Bonus Tools

• docker_get_vulnerabilities: Fetch security scan results (if available)

• docker_get_image_history: Get image build history

• docker_track_base_updates: Check if base images have updates

• docker_estimate_pull_size: Estimate the total download size for a DockerHub image tag (shows total bytes and layer breakdown).

📖 Tool API Reference

Each tool is available via the MCP protocol. Here’s a quick summary of what each tool expects and returns:

docker_search_images

Input: { query: string } Output: { results: any }

docker_get_image_details

Input: { namespace: string, repository: string } Output: { details: any }

docker_list_tags

Input: { namespace: string, repository: string } Output: { tags: string[] }

docker_get_manifest

Input: { namespace: string, repository: string, tag: string } Output: { manifest: any }

docker_analyze_layers

Input: { namespace: string, repository: string, tag: string } Output: { layers: any[], totalSize: number }

docker_compare_images

Input: { image1: { namespace, repository, tag }, image2: { namespace, repository, tag } } Output: { comparison: any }

docker_get_dockerfile

Input: { namespace: string, repository: string, tag: string } Output: { dockerfile: string | null }

docker_get_stats

Input: { namespace: string, repository: string } Output: { pull_count: number, star_count: number }

🧑‍💻 Usage Examples

This server is designed to be used by MCP-compatible clients (like Claude Desktop, Cursor, Cline, etc). You can also write your own script to send MCP requests, or use the provided tests as examples.

🛠 Troubleshooting

• Authentication errors? Double-check your .env file and DockerHub credentials.

• Rate limits? DockerHub enforces strict rate limits. If you hit them, try again later or use authenticated requests.

• Tool not working as expected? Run npm test to check for issues, and see the logs for error messages.

📚 Documentation

Setup and Configuration

• Quick Setup Guide - Step-by-step setup for Claude Desktop

• Authentication Guide - Complete authentication setup for private repositories

• Troubleshooting Guide - Common issues and solutions

API Reference

• API Documentation - Complete API reference for all 15 tools

• MCP Schema Explanation - Understanding MCP tool schemas

Technical Documentation

• Implementation Details - Architecture decisions and technical details

• Assignment Checklist - Development completion tracking

🔗 Additional Resources

• MCP TypeScript SDK

• Docker Hub API Docs

• MCP Specification

📝 License

MIT