📡 MCP Radar

wakeup595626-cmyk/mcphelm

Open-source control center for the OpenAI Secure MCP Tunnel: register local MCP servers (stdio/HTTP), run and supervise the official tunnel-client, and connect ChatGPT / Codex through a desktop app, l

⭐ 5 stars
⑂ 0 forks
TypeScript
momentum ▲ 8.3
created 2026-10-06
on radar since 2026-10-08
star trend 4 → 5 since 2026-10-08
ai-toolschatgptclicodexdesktop-appdeveloper-toolselectronmcpmcp-servermodel-context-protocolopen-sourceopenaisecure-mcp-tunneltunneltunnel-clientwindows
View on GitHub ↗Homepage ↗

About mcphelm

A local control center that connects your local MCP servers to ChatGPT / Codex through the official Secure MCP Tunnel: register servers, bring a tunnel up, and manage it from a desktop window, a local web panel or the CLI.

MCPHelm is an early 0.1.x release (the first public version was 0.1.0) and is still iterating quickly: the CLI flags and the config format may still change in incompatible ways.

Windows users can install the packaged app straight from Releases; the package is not published to npm yet, so other platforms run from source.

The OpenAI Secure MCP Tunnel lets ChatGPT / Codex reach MCP servers running on your own machine without exposing them to the public internet: the official tunnel-client opens an outbound connection from your machine, and the platform finds you by tunnel ID.

The official tool is a command-line binary. Past the first tunnel, everyday questions turn into manual work: which tunnel maps to which server, which environment variable holds the key, is the health port taken, where are the logs, is the process still alive. MCPHelm adds a control center that runs only on your machine:

- one config file that registers your local MCP servers and tunnels; - one set of commands that manages their lifecycle (start, stop, status, logs, doctor); - one local-only web panel that lays out status, health probes and logs.

It ships no license key, activation or validation logic, and it never asks you to hand your key to a third party: you create the runtime key on the OpenAI platform and keep it in your own environment variables.

From the project README.

Maintaining this server?

Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:

[![On MCP Radar](https://mcp.liqiwa.com/badge.svg)](https://mcp.liqiwa.com/s/wakeup595626-cmyk--mcphelm.html)

Related MCP servers

mcptoai/MCPtoAI

Connect AI models to your own computer and MCP tools — with device-side control.

⭐ 3Python
klarluft/gitwarren-app

A cross-platform desktop app for doing local code reviews of your own git repositories. Single user, single machine, no server, no account.

⭐ 11TypeScript
fengyincheng/ShellBridge

Give ChatGPT fast, safe, and auditable visibility into your VPS through MCP.

⭐ 4TypeScript
adityajha2005/decisionkit

Fast, cheap, probability-scored decisions for coding agents, scripts and CI. CLI + MCP server + Claude Code plugin for the OpenAI Decisions API and Jev (TypeSafe AI).

⭐ 4TypeScript
alexanderradahl/mac-developer-bridge

Give ChatGPT a real terminal on your Mac. Open-source MCP bridge for shell, files, PTY sessions, jobs, and Codex history.

⭐ 22JavaScript
Vuk97/codex-remote-pro

ChatGPT <-> Codex bridge: local coding agents ask ChatGPT for advice, and ChatGPT can inspect or steer your running Codex sessions. No API key.

⭐ 12Python

📬 Get the weekly radar in your inbox

The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.