On-Premise MCP Design Platform

A professional-grade, containerized Model Context Protocol (MCP) server platform designed for visual design tool workflows. This project provides a composable, scalable architecture for integrating multiple specialized tools into a unified AI-accessible interface.

Built for developers who want to "stay frosty" and implement cutting-edge technology patterns while maintaining production-quality standards.

🎯 Vision

Transform complex, multi-tool visual design workflows into AI-accessible services that can be operated remotely through chat interfaces. This platform serves as a proof of concept for professional MCP server architecture that can scale from hobby projects to enterprise solutions.

Core Philosophy

• Test-Driven Development: Every feature is built with tests first, ensuring reliability and maintainability

• Surgical Precision: Focused, modular components that do one thing exceptionally well

• Container-First: Built for consistent deployment across development, staging, and production

• AI-Native: Designed specifically for AI agent interaction patterns

Related MCP server: @designjs/mcp-server

🚀 Features

Current Capabilities

• Composable MCP Architecture: Multiple specialized FastMCP servers mounted under a unified FastAPI application

• Asynchronous Job Processing: Built-in support for long-running tasks with job queues and progress tracking

• Container-First Deployment: Complete Docker containerization with development and production configurations

• Professional Testing: Comprehensive test suite with near 100% coverage using pytest and TDD patterns

• Modern Python Tooling: Built with uv, pyproject.toml, and contemporary Python best practices

• Plugin Architecture: Tool servers discovered and mounted dynamically

Planned Features

• Visual Design Tool Integration: Seamless interaction with design software APIs

• Multi-Environment Support: Development, staging, and production environment configurations

• Monitoring & Observability: Built-in logging, metrics, and health checks

• Authentication & Authorization: Secure access control for production deployments

🏗️ Architecture

┌─────────────────────────────────────────┐
│             LM Studio Client            │
└─────────────────┬───────────────────────┘
│ MCP Protocol
┌─────────────────▼───────────────────────┐
│         FastAPI Main Application        │
│     ┌─────────────────────────────────┐ │
│     │    Image Processing Tools       │ │
│     │    (FastMCP Server)            │ │
│     └─────────────────────────────────┘ │
│     ┌─────────────────────────────────┐ │
│     │    File System Tools           │ │
│     │    (FastMCP Server)            │ │
│     └─────────────────────────────────┘ │
│     ┌─────────────────────────────────┐ │
│     │    Design Workflow Tools        │ │
│     │    (FastMCP Server)            │ │
│     └─────────────────────────────────┘ │
└─────────────────┬───────────────────────┘
│
┌─────────────────▼───────────────────────┐
│          Job Queue System               │
│     (Redis/RabbitMQ + Workers)         │
└─────────────────────────────────────────┘

📋 Requirements

Development Environment

• Python: 3.11+ (managed with uv)

• Docker: Latest stable version

• Container Runtime: Docker Desktop or compatible

Target Deployment

• Windows PC: NVIDIA GPU-enabled workstation for design tools

• macOS Client: i.e. M4 Max with 48GB RAM running LM Studio

🚀 Quick Start

1. Clone and Setup

# Clone the repository

git clone https://github.com/yourusername/mcp-design-platform.git
cd mcp-design-platform

# Create and activate virtual environment with uv

uv venv
source .venv/bin/activate  \# On Windows: .venv\Scripts\activate

# Install dependencies

uv pip install -e .

2. Development with Docker

# Build the development image

docker build -f Dockerfile.dev -t mcp-platform:dev .

# Start the Redis service with Docker Compose before running the server
docker-compose -f docker-compose.dev.yml up -d

# The MCP server will be available at http://localhost:8080

3. Configure LM Studio

Add to your mcp.json:

{
  "mcpServers": {
    "design-platform": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "--interactive",
        "-p",
        "8080:8080",
        "mcp-platform:latest"
      ]
    }
  }
}

📝 Note Management

You can retrieve note contents programmatically using read_note:

from mcp_platform import server

server.notes['welcome'] = 'hello world'
content = server.read_note('welcome')

read_note raises ValueError if the note does not exist.

The server also exposes a read-note tool to fetch a note via the MCP protocol:

await server.handle_call_tool('read-note', {'name': 'welcome'})

🧪 Testing

This project follows strict Test-Driven Development practices:

# Run all tests with coverage

pytest --cov=src --cov-report=html --cov-report=term

# Run tests in watch mode during development

pytest-watch

# Run only unit tests

pytest tests/unit/

# Run integration tests

pytest tests/integration/

Coverage Target: >95% line coverage, >90% branch coverage

📁 Project Structure

mcp-design-platform/
├── src/
│   ├── mcp_platform/
│   │   ├── main.py              \# FastAPI application entry point
│   │   ├── servers/             \# Individual MCP server modules
│   │   │   ├── image_tools.py
│   │   │   ├── file_tools.py
│   │   │   └── workflow_tools.py
│   │   ├── jobs/                \# Asynchronous job processing
│   │   │   ├── queue.py
│   │   │   ├── workers.py
│   │   │   └── tasks.py
│   │   └── config/              \# Configuration management
│   │       ├── settings.py
│   │       └── environments.py
├── tests/
│   ├── unit/                    \# Fast, isolated tests
│   ├── integration/             \# Component interaction tests
│   └── fixtures/                \# Test data and helpers
├── docker/
│   ├── Dockerfile               \# Production image
│   ├── Dockerfile.dev           \# Development image
│   └── docker-compose.yml       \# Multi-service orchestration
├── docs/                        \# Documentation
├── AGENTS.md                    \# AI development instructions
├── README.md                    \# This file
└── pyproject.toml              \# Modern Python project configuration

🗺️ Roadmap

Phase 1: Foundation (Completed)

• Project scaffolding with modern Python tooling

• Basic FastAPI + FastMCP integration

• Docker containerization

• TDD workflow establishment

• Redis job queue integration

• Basic tool implementations

Phase 2: Core Platform (In Progress)

• Complete asynchronous job processing system

• Progress tracking and status endpoints

• Comprehensive error handling and logging

• Production-ready Docker configurations

• CI/CD pipeline setup

Phase 3: Design Tool Integration (Following 6-8 weeks)

• Visual design software API integrations

• File system operation tools

• Image processing and manipulation tools

• Workflow automation capabilities

• Cross-platform compatibility testing

Phase 4: Production Features (Future)

• Authentication and authorization

• Multi-tenant support

• Monitoring and observability

• Performance optimization

• Enterprise deployment guides

🤝 Contributing

This project is designed to be developed collaboratively with AI agents using the patterns described in AGENTS.md.

Development Workflow

1. Issues First: All work begins with a GitHub issue describing the requirement

2. Test-Driven: Write failing tests before implementing features

3. Small Increments: Keep changes focused and atomic

4. Container Testing: All tests must pass in containerized environments

Getting Started

1. Review AGENTS.md for AI development guidelines

2. Check open issues for current priorities

3. Follow the TDD cycle: Red → Green → Refactor

4. Submit pull requests with comprehensive tests

📄 License

MIT License - see LICENSE for details.

🙏 Acknowledgments

Built with inspiration from:

• FastMCP - The foundation for MCP server development

• Model Context Protocol - The protocol specification

• Professional software development practices from the Python and containerization communities

This project represents a commitment to professional-grade software development practices while exploring cutting-edge AI integration patterns. It's designed to be both a learning vehicle and a foundation for production systems.