Natural language interface for installing MCP servers to Claude Code, plus context data and reference materials for AI agents to generate valid MCP commands.

Install MCP servers using natural language:

# Install the tool
./install.sh
# Use natural language to install MCP servers
claudemcp "install notion"
claudemcp "add github integration"
claudemcp --execute "I need stripe for payments"
claudemcp -e "connect to postgres database at localhost"

The claudemcp tool uses Claude AI with the context data in this repo to generate valid claude mcp add commands from your natural language descriptions.

Claude-Code-MCP-Command-Generator/
├── claudemcp               # CLI tool for natural language MCP installation
├── install.sh             # Installation script
├── context-data/          # Structured reference materials for AI agents
│   ├── README.md         # Guide to using the context files
│   ├── ai-reference.md   # Core command syntax and rules
│   ├── server-catalog.md # Complete catalog of MCP servers
│   └── command-examples.md # Practical usage patterns
└── reference.md          # Original full documentation (for humans)

# Clone this repository
git clone https://github.com/yourusername/Claude-Code-MCP-Command-Generator.git
cd Claude-Code-MCP-Command-Generator
# Run the installer
./install.sh

This installs claudemcp to ~/.local/bin (or wherever $INSTALL_DIR points).

• Claude Code CLI must be installed and available in your PATH
• Bash shell

claudemcp [OPTIONS] "<description>"
OPTIONS:
    -e, --execute       Execute the generated command immediately
    -s, --scope SCOPE   Specify scope: local, project, or user (default: local)
    -h, --help          Show help message

# Generate command (shows command without executing)
claudemcp "install notion"
# Output: claude mcp add --transport http notion https://mcp.notion.com/mcp
# Generate and execute immediately
claudemcp --execute "add github integration"
# With specific scope
claudemcp --scope project "add sentry for error monitoring"
# Complex database connection
claudemcp -e "connect to postgres at prod.db.com port 5432 database analytics"
# API-based services
claudemcp "install airtable with my api key"
# Output: claude mcp add --transport stdio airtable --env AIRTABLE_API_KEY=YOUR_KEY -- npx -y airtable-mcp-server

1. You describe what MCP server you want in natural language
2. claudemcp feeds your request + context data to Claude CLI
3. Claude generates the proper claude mcp add command with correct syntax
4. The command is displayed (and optionally executed)

The tool uses the curated context data in this repository to ensure Claude understands:

• Correct command syntax for different transport types
• Available MCP servers and their URLs
• Authentication requirements
• Platform-specific considerations

To generate valid Claude Code MCP commands:

1. Understand syntax: Read context-data/ai-reference.md for command structure
2. Find server: Look up server details in context-data/server-catalog.md
3. See examples: Reference context-data/command-examples.md for patterns

Compact reference containing:

• Command structure for all transport types (HTTP, SSE, Stdio)
• Management commands
• Configuration scopes and precedence
• Authentication types
• Environment variables
• Platform-specific notes
• Command generation rules for AI agents

Complete catalog organized by category:

• Development & Testing Tools (Sentry, Socket, Hugging Face, Jam)
• Project Management (Asana, Atlassian, Linear, Notion, etc.)
• Databases (Airtable, Daloopa, HubSpot)
• Payments & Commerce (PayPal, Plaid, Square, Stripe)
• Design & Media (Figma, Cloudinary, invideo, Canva)
• Infrastructure & DevOps (Cloudflare, Netlify, Vercel)
• Automation & Integration (Workato, Zapier)

Each entry includes URL, transport type, auth method, and ready-to-use command.

Practical examples including:

• Basic command templates
• Scope usage examples
• Authentication workflows
• Real-world use cases
• Platform-specific examples
• Troubleshooting examples

Original comprehensive documentation from Claude Code docs. Contains the complete source material including React components and full explanations. Use this for human reading or deep reference.

# HTTP Server (Recommended)
claude mcp add --transport http <name> <url>
# SSE Server (Deprecated)
claude mcp add --transport sse <name> <url>
# Stdio Server (Local)
claude mcp add --transport stdio <name> [--env KEY=value] -- <command> [args...]

claude mcp add --transport http notion https://mcp.notion.com/mcp
# Then authenticate in Claude Code with: /mcp

claude mcp add --transport stdio airtable --env AIRTABLE_API_KEY=YOUR_KEY -- npx -y airtable-mcp-server

claude mcp add --transport stdio db -- npx -y @bytebase/dbhub --dsn "postgresql://user:pass@host:5432/database"

Total context data: ~25 KB (optimized for AI agent parsing)

• ai-reference.md: ~5 KB
• server-catalog.md: ~8 KB
• command-examples.md: ~12 KB
• reference.md: ~40 KB (original full docs)

This repository provides structured, chunked context data optimized for AI agents to:

• Parse command syntax accurately
• Generate valid MCP commands
• Look up server details quickly
• Reference usage patterns
• Validate command structure

Derived from official Claude Code MCP documentation at https://docs.claude.com/en/docs/claude-code/mcp

Documentation content follows Claude Code documentation licensing.