MCP Quick Setup Tool - Quick Start Guide

🚀 Get up and running with MCP servers in 5 minutes!

Installation (Choose Your Platform)

🐧 Linux & 🍎 macOS

# One-line installer
curl -fsSL https://raw.githubusercontent.com/ModelContextKit/modelctx/main/INSTALL.sh | bash

# Or manual clone and install
git clone https://github.com/ModelContextKit/modelctx.git
cd mcp-setup-tools
./INSTALL.sh

🪟 Windows

Option 1: PowerShell (Recommended)

# Download and run
irm https://raw.githubusercontent.com/ModelContextKit/modelctx/main/install.ps1 | iex

# Or from cloned repository
git clone https://github.com/ModelContextKit/modelctx.git
cd mcp-setup-tools
.\install.ps1

Option 2: Command Prompt

git clone https://github.com/ModelContextKit/modelctx.git
cd mcp-setup-tools
install.bat

Your First MCP Server

Step 1: Activate Environment

# Linux/macOS
source ~/mcp-setup-tool/activate_mcp.sh

# Windows PowerShell
. $env:USERPROFILE\mcp-setup-tool\activate_mcp.ps1

Step 2: Create Your First Server

# Interactive wizard (recommended for beginners)
mcp-setup wizard

# Or create directly
mcp-setup create my-first-server --backend database

Step 3: Configure and Test

# Navigate to your project
cd my-first-server

# Configure environment (edit with your database details)
cp .env.template .env
# Edit .env file with your settings

# Test the server
python server.py

Common Use Cases

📊 Database Integration

mcp-setup create sales-db --backend database
# Configure with PostgreSQL, MySQL, or SQLite
# Provides tools: execute_query, get_table_schema, list_tables

🌐 REST API Integration

mcp-setup create api-client --backend api
# Connect to any REST API with authentication
# Provides tools: api_request, get_api_status

📁 File System Access

mcp-setup create file-manager --backend filesystem
# Safe file operations with access controls
# Provides tools: read_file, write_file, list_directory

🕷️ Web Scraping

mcp-setup create web-scraper --backend webscraper
# Scrape websites with rate limiting and respect for robots.txt
# Provides tools: scrape_url, extract_links, take_screenshot

Integration with Claude Desktop

After creating your MCP server:

1. Locate Claude Desktop Config

# Linux
~/.config/Claude/claude_desktop_config.json

# macOS
~/Library/Application Support/Claude/claude_desktop_config.json

# Windows
%APPDATA%\Claude\claude_desktop_config.json

2. Add Your Server

Copy the configuration from your project's config/claude_desktop_config.json:

{
  "mcpServers": {
    "my-first-server": {
      "command": "python",
      "args": ["/path/to/my-first-server/server.py"],
      "env": {
        "DATABASE_URL": "postgresql://user:pass@localhost/db"
      }
    }
  }
}

3. Restart Claude Desktop

Your MCP server will now be available in Claude Desktop!

Testing Your Server

With MCP Inspector (Recommended)

# Install Node.js if not already installed
# Then test your server:
npx @modelcontextprotocol/inspector python server.py

With Built-in Tests

# Run the generated test suite
cd my-first-server
python -m pytest tests/

Manual Testing

# Start server and test manually
python server.py
# Server runs on stdio - send MCP protocol messages

Next Steps

🔧 Customize Your Server

Edit config/config.yaml for server settings
Modify server.py to add custom logic
Add new tools by extending the backend class

📚 Learn More

Read the full Setup Guide for detailed instructions
Check out API Documentation for advanced usage
Explore Example Projects for inspiration

🚀 Deploy Your Server

# Local deployment
python server.py

# Docker deployment (if Dockerfile generated)
docker build -t my-mcp-server .
docker run my-mcp-server

# Cloud deployment
mcp-setup deploy my-server --target cloud

Troubleshooting

❌ Common Issues

Python Version Error

# Check Python version
python --version
# Must be 3.8 or higher

# Linux: sudo apt install python3.9
# macOS: brew install python@3.11
# Windows: Download from python.org

Permission Errors

# Don't use sudo with pip in virtual environments
# Recreate virtual environment if needed:
rm -rf mcp-env
python -m venv mcp-env
source mcp-env/bin/activate  # Linux/macOS

Import Errors

# Ensure virtual environment is activated
# Check if packages are installed:
pip list | grep mcp

Database Connection Issues

# Check database URL format:
# PostgreSQL: postgresql://user:password@host:port/database
# MySQL: mysql://user:password@host:port/database
# SQLite: sqlite:///path/to/database.db

# Test connection:
python -c "import psycopg2; print('PostgreSQL OK')"

🆘 Get Help

📖 Detailed Setup Guide
🐛 Report Issues
💬 Community Discussions
📧 Support: Create an issue for help

Available Commands

# List all available backend types
mcp-setup list

# Create new project
mcp-setup create <name> --backend <type>

# Interactive project creation
mcp-setup wizard

# Manage templates
mcp-setup templates --list

# Generate documentation
mcp-setup docs <project-path>

# Deploy project (if configured)
mcp-setup deploy <project-name>

# Show help
mcp-setup --help

Examples by Backend Type

Database Backend

mcp-setup create inventory-db --backend database
# Use case: Inventory management system
# Tools: Query products, update stock, generate reports

API Backend

mcp-setup create weather-api --backend api
# Use case: Weather data integration
# Tools: Get current weather, forecasts, alerts

Filesystem Backend

mcp-setup create document-manager --backend filesystem
# Use case: Document management system
# Tools: Search documents, read content, organize files

Web Scraper Backend

mcp-setup create news-scraper --backend webscraper
# Use case: News aggregation
# Tools: Scrape articles, extract headlines, get summaries

Ready to build amazing AI integrations? Start with mcp-setup wizard! 🚀

FilesExpand file tree

QUICK_START.md

Latest commit

History