mcp-on-vercel
About
A template Model Context Protocol (MCP) server built with the TypeScript SDK, designed for deployment on Vercel. It lets developers define custom tools, prompts, and resources and test them locally or on Vercel’s infrastructure.
Details
- Author
- xarical
- Downloads
- 348
- Categories
- Cloud Service
Jump to
- Ready‑to‑deploy MCP server template for Vercel
- Supports Server‑Sent Events (SSE) and streamable HTTP transports
- Includes test client scripts for both transports
- Uses standard TypeScript SDK for tool/resource definitions
- Environment variables accessible via process.env
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
mcp-on-vercelCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install dependencies with npm install, then run the dev server with vercel dev. Edit api/server.ts to add your tools, prompts, and resources. Test using the provided SSE or streamable HTTP client scripts.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp-on-vercel": {
"mcp-on-vercel": {
"command": "node",
"args": [
"scripts/test-sse-client.mjs",
"YOUR_MCP_SERVER_URL"
]
}
}
}
}
McpServers
{
"mcp-on-vercel": {
"command": "node",
"args": [
"scripts/test-sse-client.mjs",
"YOUR_MCP_SERVER_URL"
]
}
}
mcp-on-vercel
A template Model Context Protocol (MCP) server for deployment on Vercel.
Update api/server.ts with your tools, prompts, and resources following the MCP TypeScript SDK documentation
Running locally
Run dev server:
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.

