MCP Sandbox

🎮 Live Demo | 📦 Demo Repository | 📝 Blog Post

Overview

MCP Sandbox enables you to run Model Context Protocol (MCP) servers directly in the browser using E2B's sandbox environment. This package makes it easy to integrate AI assistants with tools and data sources without requiring any local setup.

Installation

npm install @netglade/mcp-sandbox

Quick Start

import { startMcpSandbox } from '@netglade/mcp-sandbox';

// Start the MCP sandbox
const mcpSandbox = await startMcpSandbox({
  command: 'npx -y @modelcontextprotocol/server-brave-search',
  apiKey: 'e2b_****',
});

// Get the MCP server URL to connect your AI assistant
const mcpUrl = mcpSandbox.getUrl();
console.log("MCP server URL:", mcpUrl);

How It Works

1. MCP Server Execution: Uses supergateway to run stdio-based MCP servers over SSE (Server-Sent Events)
2. Sandbox Environment: Runs in E2B's secure sandbox environment, providing isolated execution
3. Tool Integration: Seamlessly connects AI assistants with real-world tools and data sources
4. Browser Support: Everything runs directly in the browser without local dependencies

Examples and Resources

• 🎮 Live Demo - Try it directly in your browser
• 📦 Demo Repository - Full example implementation
• 📝 Blog Post - Learn about the project's journey

Contributing

We welcome contributions! Feel free to open issues and pull requests.

License

MIT