🚀 MCP Smart Scheduler & Todo App (Vercel & Remote Ready)

A full-stack Scheduler & Todo web application powered by Model Context Protocol (MCP). Works both locally and remotely on Vercel with ChatGPT, Claude Code, Cursor, and Antigravity.

📊 UI Table Format

☁️ How to Deploy to Vercel

Option 1: Vercel CLI (Fastest)

1. Install Vercel CLI if not installed:

   npm i -g vercel

2. Deploy directly:

   vercel

   Follow the prompts and your app will be live at https://<your-project-name>.vercel.app!

Option 2: GitHub Repository Integration

1. Push this project code to your GitHub Repository.

2. Go to Vercel Dashboard.

3. Import the repository and click Deploy. Vercel will automatically detect vercel.json and build the app.

🤖 Connecting Remote MCP (Vercel URL) to ChatGPT / Claude Code / Cursor

Once deployed to Vercel, your Remote MCP endpoints are:

• SSE Endpoint: https://<your-app-name>.vercel.app/api/mcp/sse

• JSON-RPC Endpoint: https://<your-app-name>.vercel.app/api/mcp

1. Claude Desktop / Cursor / Claude Code (mcp-config.json)

Add this configuration to your local MCP settings file:

{
  "mcpServers": {
    "scheduler-todo-remote": {
      "url": "https://<your-app-name>.vercel.app/api/mcp/sse"
    }
  }
}

2. Testing via ChatGPT (Custom Actions / Remote HTTP)

ChatGPT Remote Actions can make POST requests to https://<your-app-name>.vercel.app/api/mcp with standard MCP JSON-RPC payloads:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "add_reminder",
    "arguments": {
      "time": "12:00 PM",
      "work": "Meeting with Team"
    }
  }
}