Cube A Labs MCP Server

Model Context Protocol server that exposes Cube A Labs skills to AI agents like Claude Code.

What is this?

This MCP server allows AI coding agents to automatically discover and use Cube A Labs skills:

• Favicon Generator - Complete favicon sets with PWA support
• Cube UI Generator - Framework-aware UI component generation
• Sitemap Generator - SEO-optimized XML sitemaps

Installation

Option 1: NPM (Recommended)

npm install -g @cubealabs/mcp-server

Option 2: NPX (No Install)

Configure Claude Code to use npx:

{
  "mcpServers": {
    "cubealabs": {
      "command": "npx",
      "args": ["-y", "@cubealabs/mcp-server"]
    }
  }
}

Configuration

Claude Code

Add to your Claude Code settings (~/.config/claude/config.json or use /mcp config):

{
  "mcpServers": {
    "cubealabs": {
      "command": "cubealabs-mcp"
    }
  }
}

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "cubealabs": {
      "command": "cubealabs-mcp"
    }
  }
}

Other MCP Clients

Any MCP-compatible client can connect using stdio transport:

cubealabs-mcp

Available Tools

The server exposes three tools:

cubealabs_favicon

Generate complete favicon sets from any logo with PWA support, all icon sizes, and framework integration.

Use cases:

• Setting up favicons for new projects
• Adding PWA icons to existing apps
• Generating dark mode favicon variants
• Framework-specific integration (Next.js, React, Vue, etc.)

cubealabs_cube_ui

Generate UI components with automatic framework detection.

Use cases:

• Creating new components (cards, buttons, forms)
• Building page layouts
• Generating responsive, accessible components
• Framework-specific code generation

cubealabs_sitemap

Generate SEO-optimized XML sitemaps with route discovery.

Use cases:

• Creating sitemaps for Google Search Console
• Automatic route discovery from file system
• Handling dynamic routes
• Framework-specific sitemap generation

How It Works

1. AI agent requests available tools from the MCP server
2. Server responds with list of Cube A Labs skills
3. AI agent selects appropriate skill based on user request
4. Server provides skill instructions to the AI agent
5. AI agent executes skill following the instructions

Development

Build the server:

npm install
npm run build

Test locally:

npm run dev

Links

• GitHub Repository
• NPM Package
• MCP Documentation

License

MIT © Cube A Labs