MCP Servers for CS Experimentation Workshop

A collection of MCP servers designed for rapid prototyping in CS experimentation workshops.

GitHub

MCP Servers for CS Experimentation Workshop

Achieve "5-minute magic" - go from concept to working experiment in under 5 minutes using AI-collaborative development with Model Context Protocol servers.

Overview

This project implements a comprehensive Model Context Protocol (MCP) server ecosystem designed specifically for rapid CS experimentation. Built for Mac Studio M3 Ultra optimization, it provides seamless integration between AI assistants and development tools.

Features

  • Sub-5-minute Setup: From zero to fully configured development environment
  • Multi-Tier Architecture: Essential, database, and AI collaboration servers
  • Docker Isolation: Secure, containerized environments for each experiment
  • AI Memory: Persistent context across sessions
  • Multi-Database Support: PostgreSQL, Redis, MongoDB, Neo4j out of the box
  • GitHub Integration: Automated repository management and CI/CD
  • TypeScript Build: Clean compilation with full type safety ✅

Quick Start

Option 1: MCP Starter Kit (Recommended for New Users)

# Get the starter kit for a 5-minute setup
cd mcp-starter-kit/
./setup.sh

# Or with 1Password integration:
./setup-with-1password.sh

# Restart Claude Desktop and you're ready! 🚀

Option 2: Full Workshop Setup

# Clone the repository
git clone https://github.com/slcc2c/mcp-workshop-servers.git
cd mcp-workshop-servers

# Install dependencies
npm install

# Set up 1Password integration (recommended)
./scripts/add-database-connections.sh

# Start all MCP servers
npm run mcp:start

📚 Read the Quick Start Guide for detailed setup instructions.

Attaching Your Project

Quick Attach (2 minutes)

# Run the interactive attachment script
./scripts/attach-project.sh

# Follow the prompts to:
# 1. Select your project directory
# 2. Choose a project name
# 3. Configure filesystem + memory access

# Restart Claude Desktop and start coding!

📎 Read the Project Attachment Guide for detailed instructions and advanced configurations.

Project Templates

Use our pre-configured templates for common project types:

  • Web App: React + Node.js full-stack template
  • API: RESTful API with authentication and docs
  • Data Science: Python ML project with Jupyter
  • Mobile: React Native cross-platform template
# View available templates
ls templates/

# Use a template
cp -r templates/web-app/* /path/to/your/project/

Architecture

Tier 1: Essential Foundation

  • GitHub Server: Repository management, issue tracking, PR automation
  • Filesystem Server: Secure file operations with configurable access
  • Docker Server: Container management and isolated environments
  • Memory Server: Persistent AI context and knowledge graphs

Tier 2: Database & Services

  • PostgreSQL/Supabase: Full database management with real-time features
  • Redis: Caching, sessions, and real-time data structures
  • MongoDB: Document store with Atlas integration
  • Neo4j: Graph database for relationship-heavy experiments
  • Kubernetes: Service orchestration and deployment

Tier 3: AI Collaboration

  • Jupyter Integration: Live notebook execution
  • FastMCP Framework: Rapid tool creation
  • Code Analysis: Automated documentation and insights

Configuration

Configure Claude Desktop by adding to claude_config.json:

{
  "mcpServers": {
    "mcp-workshop": {
      "command": "node",
      "args": ["/path/to/mcp-workshop-servers/dist/index.js"],
      "env": {
        "GITHUB_TOKEN": "${GITHUB_TOKEN}",
        "WORKSHOP_MODE": "true"
      }
    }
  }
}

Documentation

Documentation

Contributing

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

License

MIT License - see LICENSE for details.

Acknowledgments

Built on the Model Context Protocol by Anthropic, leveraging the vibrant MCP ecosystem and community servers.

Máy chủ liên quan