yogocomfyorg/Comfy.org_MCP-Server

🎨 ComfyUI MCP Server

A comprehensive Model Context Protocol (MCP) server for ComfyUI with automatic agentic controlling capabilities, intelligent workflow orchestration

🌟 Key Highlights

• 🤖 Fully Autonomous ComfyUI Control - Complete API automation with intelligent decision making
• 🎭 Advanced Workflow Orchestration - Chain complex workflows with dependencies and error handling
• 🖼️ Intelligent Image Routing - Automatic image bridging between workflows
• 🔧 Smart Custom Node Management - Automated installation with dependency resolution
• 📦 One-Click ComfyUI Installation - Complete setup with CUDA optimization options
• 🚀 Performance Optimization - Triton and SageAttention integration
• 🔍 Real-time Monitoring - System health and performance tracking

🎯 Core Capabilities

🤖 Autonomous AI Control

• Intelligent Workflow Management: AI-driven workflow selection and optimization
• Automatic Error Recovery: Smart error detection and resolution
• Adaptive Performance Tuning: Dynamic optimization based on system capabilities

🎭 Advanced Workflow System

• Workflow Orchestration: Chain multiple workflows with complex dependencies
• Parameter Injection: Dynamic parameter modification for any workflow type
• Image Routing: Intelligent bridging of images between workflow stages
• Batch Processing: Execute multiple workflows with parallel processing

🔧 Complete ComfyUI Management

• Model Download & Management: HuggingFace, Civitai, and custom model sources
• Custom Node Ecosystem: Automated installation, updates, and dependency management
• Configuration Management: Dynamic ComfyUI settings and optimization
• System Integration: Terminal operations, file management, and monitoring

🚀 Performance & Optimization

• CUDA Optimization: Multi-version CUDA support with automatic selection
• Triton Integration: GPU kernel optimization for enhanced performance
• SageAttention: Memory-efficient attention mechanisms
• Resource Monitoring: Real-time CPU, RAM, and GPU utilization tracking

🚀 Quick Start

Option 1: Complete Installation (Recommended)

🎯 One-Click ComfyUI Setup with MCP Server

1. Clone the repository:

   git clone https://github.com/your-org/comfyui-mcp-server.git
   cd comfyui-mcp-server
   

2. Install ComfyUI with optimization options:

   # Run the comprehensive installer
   sandbox/Install_ComfyUI_Complete.bat
   

   The installer will guide you through:

   • CUDA Version Selection (11.8, 12.1, 12.4, 12.6, or CPU-only)
   • Triton & SageAttention optimization options
   • Build Tools validation and setup
   • Automatic Environment configuration

3. Install MCP Server dependencies:

   npm install
   npm run build
   

4. Start the MCP Server:

   npm start
   # or
   start-comfyui-mcp.bat
   

Option 2: MCP Server Only

If you already have ComfyUI installed:

1. Clone and setup:

   git clone https://github.com/your-org/comfyui-mcp-server.git
   cd comfyui-mcp-server
   npm install
   npm run build
   

2. Configure ComfyUI path in augment-mcp-config.json

3. Start the server:

   npm start
   

📋 System Requirements

Minimum Requirements

• Node.js: 18.0.0 or higher
• Python: 3.10+ (for ComfyUI)
• RAM: 8GB minimum, 16GB+ recommended
• Storage: 20GB+ free space for models

Recommended for Optimal Performance

• GPU: NVIDIA RTX 3060+ with 8GB+ VRAM
• CUDA: 12.6 (latest) or 11.8+ (compatible)
• RAM: 32GB+ for large models
• Storage: SSD with 100GB+ free space
• Build Tools: Visual Studio 2019/2022 (for Triton/SageAttention)

🎮 Usage Guide

🚀 Starting the System

Method 1: Complete System Startup

# Start ComfyUI instance (if using our installer)
sandbox/ComfyUI_cuda126/Launch_ComfyUI_CUDA126.bat

# In another terminal, start MCP Server
npm start
# or
start-comfyui-mcp.bat

Method 2: Direct MCP Server

npm start
# or
node build/index.js

🎛️ Server Configuration

Environment Variables

Create a .env file in the project root:

# ComfyUI Configuration
COMFYUI_PATH=C:/path/to/ComfyUI
COMFYUI_URL=http://127.0.0.1:8188
COMFYUI_MODELS_PATH=C:/path/to/ComfyUI/models

# Performance Settings
MAX_CONCURRENT_WORKFLOWS=3
WORKFLOW_TIMEOUT=300000
IMAGE_ROUTING_CONFIDENCE=0.7

# Monitoring
ENABLE_PERFORMANCE_MONITORING=true
LOG_LEVEL=info

# DGM Integration
ENABLE_DGM_OPTIMIZATION=true
RECURSIVE_IMPROVEMENT_CYCLES=5

Advanced Configuration

Modify src/config.ts for advanced settings:

export const config = {
  comfyui: {
    defaultUrl: 'http://127.0.0.1:8188',
    timeout: 30000,
    retryAttempts: 3
  },
  workflows: {
    directory: './workflows',
    maxConcurrent: 3,
    defaultParameters: {
      batch_size: 1,
      steps: 20,
      cfg: 7.0
    }
  },
  optimization: {
    enableTriton: true,
    enableSageAttention: true,
    autoOptimize: true
  }
};

🛠️ Available Tools

🎯 Core Workflow Management

🖼️ Intelligent Image Routing

🔧 Custom Node Ecosystem

📦 Model Management

🚀 ComfyUI Server Management

🔍 System Monitoring & Diagnostics

🎛️ Configuration & Settings

🔄 Batch Operations & Automation

🎯 Pro Tip: Use workflow chains with image routing for complete automation pipelines. The system can automatically generate images, route them through upscaling, apply effects, and organize outputs - all with a single command!

🏗️ Architecture

🔧 Modular Design

The server follows a clean, modular architecture for maximum maintainability and extensibility:

📁 comfyui-mcp-server/
├── 📁 src/
│   ├── 📄 index.ts                 # 🚀 Main MCP server entry point
│   ├── 📁 core/
│   │   ├── 📄 server.ts            # Core MCP server implementation
│   │   ├── 📄 config.ts            # Configuration management
│   │   └── 📄 utils.ts             # Shared utilities
│   └── 📁 tools/
│       ├── 📄 workflowExecution.ts # 🎭 Workflow orchestration & execution
│       ├── 📄 imageRouting.ts      # 🖼️ Intelligent image routing
│       ├── 📄 customNodes.ts       # 🔧 Custom node management
│       ├── 📄 modelManagement.ts   # 📦 Model download & organization
│       ├── 📄 serverManagement.ts  # 🚀 ComfyUI server control
│       ├── 📄 systemMonitoring.ts  # 🔍 Performance monitoring
│       ├── 📄 fileOperations.ts    # 📁 File & directory operations
│       └── 📄 batchOperations.ts   # 🔄 Batch processing
├── 📁 build/                       # 🏗️ Compiled JavaScript output
├── 📁 workflows/                   # 🎭 Workflow templates
├── 📁 sandbox/                     # 📦 Installation scripts
└── 📁 state/                       # 💾 Runtime state management

🔄 Data Flow Architecture

graph TD
    A[MCP Client] --> B[MCP Server]
    B --> C[Tool Router]
    C --> D[Workflow Engine]
    C --> E[Image Router]
    C --> F[Node Manager]
    C --> G[Model Manager]

    D --> H[ComfyUI API]
    E --> H
    F --> I[Git Operations]
    G --> J[HuggingFace/Civitai]

    H --> K[ComfyUI Instance]
    K --> L[Generated Images]
    L --> E

🧠 Intelligent Systems

Image Routing Intelligence

Image Detection → Compatibility Analysis → Confidence Scoring → Route Selection → Execution

🛠️ Development

🚀 Quick Development Setup

# Clone and setup
git clone https://github.com/your-org/comfyui-mcp-server.git
cd comfyui-mcp-server

# Install dependencies
npm install

# Development mode with hot reload
npm run dev

# Build for production
npm run build

# Run tests
npm test

# Lint and format
npm run lint
npm run format

🔧 Development Scripts

🧪 Testing Strategy

# Unit tests
npm run test:unit

# Integration tests
npm run test:integration

# End-to-end tests
npm run test:e2e

# Coverage report
npm run test:coverage

🛡️ Error Handling & Reliability

🔒 Comprehensive Error Management

• Graceful Degradation: System continues operating even when components fail
• Detailed Logging: Comprehensive error tracking with context
• Automatic Recovery: Self-healing mechanisms for common issues
• Performance Monitoring: Real-time system health tracking
• Timeout Handling: Prevents hanging operations
• Resource Cleanup: Automatic cleanup of failed operations

🔍 Monitoring & Diagnostics

// Built-in health monitoring
{
  "system": {
    "status": "healthy",
    "uptime": "2h 34m",
    "memory": "1.2GB / 16GB",
    "cpu": "15%",
    "gpu": "45%"
  },
  "comfyui": {
    "status": "running",
    "queue": 2,
    "processing": true
  },
  "workflows": {
    "completed": 156,
    "failed": 3,
    "success_rate": "98.1%"
  }
}

🚀 Performance Optimization

⚡ CUDA & GPU Acceleration

The installation script provides multiple CUDA options for optimal performance:

🔥 Advanced Optimizations

• Triton Integration: GPU kernel optimization for 20-40% performance boost
• SageAttention: Memory-efficient attention for large models
• Automatic Mixed Precision: Reduces memory usage while maintaining quality
• Dynamic Batching: Optimizes batch sizes based on available VRAM

📚 Examples & Use Cases

🎨 Creative Workflows

// Generate and enhance artwork pipeline
{
  "name": "artistic_enhancement_pipeline",
  "steps": [
    {
      "id": "base_generation",
      "workflow": "text2img_api",
      "parameters": {
        "prompt": "digital art, concept art style",
        "steps": 30,
        "cfg": 8.0
      }
    },
    {
      "id": "upscale_enhance",
      "workflow": "upscaling_api",
      "dependencies": ["base_generation"],
      "routeFromPrevious": true
    }
  ]
}

📸 Photography Enhancement

// Photo restoration and enhancement
{
  "workflow": "photo_enhancement",
  "batch_processing": true,
  "input_directory": "./photos",
  "output_directory": "./enhanced",
  "parameters": {
    "denoise_strength": 0.7,
    "upscale_factor": 2.0,
    "color_correction": true
  }
}

🎬 Video Frame Processing

// Video frame extraction and processing
{
  "pipeline": "video_frame_enhancement",
  "input": "video.mp4",
  "frame_extraction": {
    "fps": 24,
    "quality": "high"
  },
  "processing": {
    "workflow": "frame_enhancement_api",
    "batch_size": 8
  }
}

🤝 Contributing

🔧 Development Contribution

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Implement your changes with tests
4. Ensure all tests pass: npm test
5. Submit a pull request with detailed description

📝 Documentation Contribution

• Improve README sections
• Add workflow examples
• Create tutorial content
• Update API documentation

🐛 Bug Reports

Please include:

• System specifications
• ComfyUI version
• Error logs and stack traces
• Steps to reproduce

💡 Feature Requests

• Describe the use case
• Explain expected behavior
• Provide implementation suggestions

📄 License

MIT License - see file for details.

This project is open source and welcomes contributions from the community.

🆘 Support & Community

📞 Getting Help

• 📖 Documentation: Check this README and tool documentation
• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions
• 📧 Email: support@your-org.com

🔧 Troubleshooting

Common Issues

Debug Mode

# Enable verbose logging
DEBUG=comfyui:* npm start

# Check system health
npm run health-check

# Validate installation
npm run validate-setup

🌟 Community

Join our growing community of AI artists, developers, and researchers using ComfyUI MCP Server for creative and technical projects.

⭐ Star this repository if you find it useful!

🔄 Watch for updates and new features

🤝 Contribute to make it even better