Docker Hub MCP Server

A comprehensive Model Context Protocol (MCP) server that provides seamless integration with Docker Hub, enabling AI assistants to search, analyze, and manage Docker images through standardized MCP tools.

🚀 Features

Core Capabilities

• Image Search & Discovery: Search Docker Hub with advanced filtering options

• Detailed Image Analysis: Get comprehensive repository information and statistics

• Tag Management: List and analyze all available tags for repositories

• Layer Analysis: Deep dive into image layers, sizes, and optimization opportunities

• Image Comparison: Compare two images for differences in layers, sizes, and composition

• Dockerfile Retrieval: Attempt to retrieve Dockerfile content when available

• Statistics & Metrics: Download counts, star ratings, and popularity analytics

Advanced Features

• Security Scanning: Vulnerability analysis and security recommendations

• Build History: Detailed layer-by-layer build process analysis

• Pull Size Estimation: Calculate actual download sizes considering caching and compression

• Performance Optimization: Layer deduplication and size optimization insights

Technical Excellence

• Production Ready: Built with TypeScript, comprehensive error handling, and rate limiting

• Smart Caching: Intelligent caching system to minimize API calls and improve performance

• Authentication Support: Secure credential management for both public and private registries

• MCP Standard Compliance: Fully compatible with popular MCP clients (Claude Desktop, Cursor, Cline)

• Multi-Transport Support: Both stdio (for MCP clients) and HTTP (for web access) transports

• Interactive Setup: Configuration wizard for easy initial setup

• Export Capabilities: Multiple export formats including CSV, SARIF, and dependency trees

• Enhanced Security: CVE cross-referencing and comprehensive security policy validation

Related MCP server: Docker Explorer

📦 Installation

Prerequisites

• Node.js 18+

• npm or yarn

Quick Start

1. Clone the repository

   git clone <repository-url>
   cd mcp-docker

2. Install dependencies

   npm install

3. Set up configuration (Interactive Wizard)

   npm run setup

   Or manually set up environment variables:

   cp env.example .env
   # Edit .env with your Docker Hub credentials and preferences

4. Build the project

   npm run build

5. Start the server

   npm start

🔧 Configuration

Environment Variables

Create a .env file from the provided template:

# Docker Hub Authentication (Optional for public images)
DOCKERHUB_USERNAME=your_dockerhub_username
DOCKERHUB_PASSWORD=your_dockerhub_password
# OR use access token instead
DOCKERHUB_ACCESS_TOKEN=your_dockerhub_access_token

# Private Registry Support (Optional)
PRIVATE_REGISTRY_URL=https://your-private-registry.com
PRIVATE_REGISTRY_USERNAME=your_private_username
PRIVATE_REGISTRY_PASSWORD=your_private_password

# Server Configuration
MCP_SERVER_NAME=dockerhub-mcp-server
MCP_SERVER_VERSION=1.0.0

# Transport Configuration
MCP_TRANSPORT=stdio  # 'stdio' for MCP clients, 'http' for web access
MCP_HTTP_HOST=localhost  # Only needed for HTTP transport
MCP_HTTP_PORT=3000  # Only needed for HTTP transport
MCP_CORS=true  # Enable CORS for HTTP transport

# Performance Tuning
CACHE_TTL_SECONDS=300
MAX_CACHE_SIZE=1000
DOCKERHUB_RATE_LIMIT=100
DOCKERHUB_RATE_LIMIT_WINDOW=3600

# Logging
LOG_LEVEL=info

Authentication

The server supports multiple authentication methods:

1. Anonymous Access: Works for all public Docker Hub images

2. Username/Password: Standard Docker Hub login credentials

3. Access Token: More secure, generated from Docker Hub settings

4. Private Registry: Support for custom registries

See SETUP.md for detailed authentication configuration.

🛠️ Available MCP Tools

Required Tools

Bonus Tools

📖 Usage Examples

Basic Image Search

{
  "tool": "docker_search_images",
  "arguments": {
    "query": "nginx",
    "limit": 10,
    "is_official": true
  }
}

Detailed Image Analysis

{
  "tool": "docker_get_image_details",
  "arguments": {
    "repository": "library/nginx"
  }
}

Layer Analysis for Optimization

{
  "tool": "docker_analyze_layers",
  "arguments": {
    "repository": "library/node",
    "tag": "18-alpine"
  }
}

Image Comparison

{
  "tool": "docker_compare_images",
  "arguments": {
    "repository1": "library/node",
    "tag1": "18-alpine",
    "repository2": "library/node", 
    "tag2": "18-slim"
  }
}

Security Analysis

{
  "tool": "docker_get_vulnerabilities",
  "arguments": {
    "repository": "library/ubuntu",
    "tag": "latest"
  }
}

Batch Analysis

{
  "tool": "docker_batch_image_details",
  "arguments": {
    "repositories": ["library/nginx", "library/node", "library/python"],
    "include_tags": true,
    "include_vulnerabilities": true,
    "format": "comparison",
    "export_format": "csv"
  }
}

Enhanced Security Analysis

{
  "tool": "docker_enhanced_vulnerability_analysis",
  "arguments": {
    "repository": "library/nginx",
    "tag": "latest",
    "severity_filter": ["critical", "high"],
    "include_cve_details": true,
    "security_policy": {
      "max_age_days": 180,
      "min_severity_threshold": "medium"
    },
    "export_format": "sarif"
  }
}

Data Export

{
  "tool": "docker_export_data",
  "arguments": {
    "repository": "library/node",
    "tag": "18-alpine",
    "export_type": "dependency-tree",
    "format": "tree-string"
  }
}

See EXAMPLES.md for more comprehensive usage examples and workflows.

🏗️ Development

Project Structure

src/
├── auth/           # Authentication management
├── cache/          # Caching infrastructure  
├── clients/        # Docker Hub API client
├── tools/          # MCP tool implementations
├── utils/          # Utilities (error handling, rate limiting)
├── types.ts        # TypeScript type definitions
├── config.ts       # Configuration management
├── server.ts       # MCP server implementation
└── index.ts        # Entry point

Development Commands

# Development with hot reload
npm run dev

# Build the project
npm run build

# Run tests
npm test

# Run with coverage
npm run test:coverage

# Lint code
npm run lint

# Fix linting issues
npm run lint:fix

Testing

The project includes comprehensive testing:

• Unit Tests: Individual function and class testing

• Integration Tests: Full API workflow testing

• Mock Tests: Testing with simulated Docker Hub responses

# Run all tests
npm test

# Run with coverage report
npm run test:coverage

# Run specific test file
npm test -- tools/search-images.test.ts

🔌 Integration with MCP Clients

Claude Desktop

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "dockerhub": {
      "command": "node",
      "args": ["/path/to/dockerhub-mcp-server/dist/index.js"],
      "env": {
        "DOCKERHUB_USERNAME": "your_username"
      }
    }
  }
}

Cursor

The server is compatible with Cursor's MCP integration. See the integration guide for setup instructions.

Cline

Works seamlessly with Cline's MCP support. Refer to Cline's documentation for MCP server configuration.

HTTP Transport (Web Access)

When using HTTP transport mode, the server provides web endpoints:

# Set HTTP transport in .env
MCP_TRANSPORT=http
MCP_HTTP_PORT=3000

# Start server
npm start

# Access endpoints
curl http://localhost:3000/health          # Health check
curl http://localhost:3000/info            # Server information
# MCP endpoint: http://localhost:3000/message

🐳 Docker Support

Using Docker Compose

A docker-compose.yml is provided for easy testing with a local registry:

# Start local registry and server
docker-compose up

# Test with local registry
curl -X POST http://localhost:3000/api/search -d '{"query": "nginx"}'

Building Docker Image

# Build the image
docker build -t dockerhub-mcp-server .

# Run the container
docker run -d \
  --name dockerhub-mcp \
  -e DOCKERHUB_USERNAME=your_username \
  -e DOCKERHUB_PASSWORD=your_password \
  dockerhub-mcp-server

🚨 Troubleshooting

Common Issues

Authentication Failures

• Verify credentials in .env file

• Check if 2FA is enabled (use access token instead)

• Ensure proper permissions for private repositories

Rate Limiting

• Default limits: 100 requests per hour

• Authenticated users get higher limits

• Implement exponential backoff for retries

Network Issues

• Check firewall settings

• Verify DNS resolution for registry-1.docker.io

• Consider proxy configuration if behind corporate firewall

See TROUBLESHOOTING.md for comprehensive troubleshooting guide.

📊 Performance & Optimization

Caching Strategy

• Image Metadata: Cached for 10 minutes

• Search Results: Cached for 5 minutes

• Layer Information: Cached for 30 minutes

• Vulnerability Scans: Cached for 1 hour

Rate Limiting

• Automatic rate limit detection from Docker Hub headers

• Intelligent backoff when limits are approached

• Queue management for high-volume requests

Memory Management

• LRU cache with configurable size limits

• Automatic cleanup of expired entries

• Memory usage monitoring and alerts

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines for details.

Development Setup

1. Fork the repository

2. Create a feature branch

3. Make your changes with tests

4. Ensure all tests pass

5. Submit a pull request

Code Standards

• TypeScript strict mode

• ESLint configuration provided

• 100% test coverage for new features

• Comprehensive documentation

📜 License

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

🙏 Acknowledgments

• Model Context Protocol for the MCP specification

• Docker Hub API for the comprehensive API

• TypeScript and Node.js communities for excellent tooling

📞 Support

• Issues: GitHub Issues

• Discussions: GitHub Discussions

• Documentation: Wiki

Built with ❤️ for the Docker and AI communities