FrontMCP

v1.2.1Developer Toolsstable

TypeScript-first framework for the Model Context Protocol (MCP). You write clean, typed code; FrontMCP handles the protocol, transport, DI, session/auth, and execution flow.

aiframeworkjavascriptmcpmcp-server
Share:
143
Stars
0
Downloads
0
Weekly
0/5

What is FrontMCP?

FrontMCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to typescript-first framework for the model context protocol (mcp). you write clean, typed code; frontmcp handles the protocol, transport, di, session/auth, and execution flow.

TypeScript-first framework for the Model Context Protocol (MCP). You write clean, typed code; FrontMCP handles the protocol, transport, DI, session/auth, and execution flow.

This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • TypeScript-first framework for the Model Context Protocol (M

Use Cases

Build MCP servers with TypeScript framework
Handle transport, DI, and authentication
Simplify MCP protocol implementation
agentfront

Maintainer

LicenseApache-2.0
Languagetypescript
Versionv1.2.1
UpdatedMay 15, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @frontmcp/ui

Manual Installation

npx -y @frontmcp/ui

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use FrontMCP

FrontMCP is a production-grade, TypeScript-first framework for building Model Context Protocol servers using decorators and dependency injection. It abstracts away the protocol, transport, session management, and authentication so developers write clean @Tool, @Resource, and @App classes instead of low-level MCP boilerplate. The same server code runs locally in development and deploys unchanged to Node.js, Vercel, AWS Lambda, or Cloudflare Workers. Teams adopt it to ship typed, testable MCP servers with built-in OAuth, JWKS, Streamable HTTP transport, and an official UI library.

Prerequisites

  • Node.js 24 or higher (strictly required by the framework)
  • npm or npx available in your PATH
  • An MCP-compatible client such as Claude Desktop, Cursor, or Windsurf
  • TypeScript knowledge for writing @Tool and @App decorators
  • reflect-metadata support (imported at the top of your entry file)
1

Scaffold a new FrontMCP project

Use the official CLI scaffolder to create a fully configured project with TypeScript, decorators, and the core SDK already wired up.

npx frontmcp create my-mcp-app
2

Install into an existing project

If you already have a Node.js project, add the FrontMCP SDK and Node types as dev dependencies, then run the init command to generate the server entry point and tsconfig adjustments.

npm i -D frontmcp @types/node@^24
npx frontmcp init
3

Define your first App and Tool

Create an App class with the @App decorator and add methods decorated with @Tool. Each method receives typed, Zod-validated inputs and returns typed outputs automatically surfaced to MCP clients.

import 'reflect-metadata';
import { App, Tool } from '@frontmcp/sdk';
import { z } from 'zod';

@App({ name: 'hello' })
export default class HelloApp {
  @Tool({ description: 'Say hello to a name' })
  async greet({ name }: { name: z.infer<typeof z.string()> }) {
    return { message: `Hello, ${name}!` };
  }
}
4

Bootstrap the FrontMCP server

Wire your App into the @FrontMcp decorator on the root server class. Set the HTTP port for local development.

import 'reflect-metadata';
import { FrontMcp, LogLevel } from '@frontmcp/sdk';
import HelloApp from './hello.app';

@FrontMcp({
  info: { name: 'Demo', version: '0.1.0' },
  apps: [HelloApp],
  http: { port: 3000 },
  logging: { level: LogLevel.Info },
})
export default class Server {}
5

Start the development server

Use the CLI dev command which watches for changes and restarts automatically. Alternatively build and run for production.

npx frontmcp dev
# or build for production:
npx frontmcp build && node dist/index.js
6

Connect your MCP client

Configure your MCP client (Claude Desktop, Cursor, etc.) to connect to the local Streamable HTTP endpoint exposed by FrontMCP.

FrontMCP Examples

Client configuration

Claude Desktop configuration connecting to a locally running FrontMCP server over HTTP/SSE on port 3000.

{
  "mcpServers": {
    "frontmcp-demo": {
      "command": "npx",
      "args": ["-y", "@frontmcp/ui"],
      "env": {}
    }
  }
}

Prompts to try

Example prompts once your FrontMCP server is running and its tools are exposed to your AI assistant.

- "Call the greet tool with my name and show the response"
- "List all tools available on the connected MCP server"
- "Use the hello app to generate a greeting for the user Alice"

Troubleshooting FrontMCP

Server fails to start with a 'version mismatch' error at boot

All @frontmcp/* packages must be on the same version. Run 'npm ls @frontmcp/sdk @frontmcp/cli' to compare versions, then update all to match: npm i -D @frontmcp/sdk@<version> @frontmcp/cli@<version>.

TypeScript errors about decorators or reflect-metadata not found

Ensure 'experimentalDecorators' and 'emitDecoratorMetadata' are both true in tsconfig.json, and that 'import reflect-metadata' is the very first line in your entry file before any other imports.

MCP client cannot connect to the server

Verify the server is listening on the expected port with 'curl http://localhost:3000'. Check that your firewall or antivirus is not blocking the port. For Claude Desktop, ensure the config JSON path is correct and the client has been restarted after config changes.

Frequently Asked Questions about FrontMCP

What is FrontMCP?

FrontMCP is a Model Context Protocol (MCP) server that typescript-first framework for the model context protocol (mcp). you write clean, typed code; frontmcp handles the protocol, transport, di, session/auth, and execution flow. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install FrontMCP?

Install via npm with the command: npx -y @frontmcp/ui. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with FrontMCP?

FrontMCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is FrontMCP free to use?

Yes, FrontMCP is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Developer Tools MCP Servers

Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "frontmcp": { "command": "npx", "args": ["-y", "@frontmcp/ui"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use FrontMCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides