RE:NODE

AI setup

Work on your server with AI, from facts instead of guesses

Most of what goes wrong when an AI assistant deploys something is a guess: the wrong port, the wrong folder, a start command the server never runs. The panel already knows the answers for your server. Press one button and it hands them to your assistant as a private link - with every password and token left out.

One button
Connect AI creates a private link for your server and copies it.
Any assistant
Claude, ChatGPT or anything that can open links - or as an MCP server.
Stops by itself
After 7 days unused, with a hard limit of 90 days. Nothing to choose.
No secrets
No password or token is ever shown to the assistant.

Still choosing? Give your assistant everything RE:NODE has published about app hosting, and ask it questions from facts instead of guesses.

Open the file

How it works

On the server's Console tab there is one bar: "Work on this server with AI".

  1. 1

    Pick a purpose

    Tell the panel what you are trying to do, so the assistant is briefed for that job rather than handed a description and left to guess.

  2. 2

    Press Connect AI

    The panel creates a private link for this server and copies it. There is no duration to choose.

  3. 3

    Paste it into your assistant

    Claude, ChatGPT or any assistant that can open links. The same link followed by /mcp also works as an MCP server URL.

The purposes

How long a link works

  • A link keeps working while it is being used and stops by itself after 7 days unused - 24 hours for "Manage this server".
  • Whatever happens, it stops after 90 days - 30 for "Manage this server".
  • Making a new link for the same purpose replaces the old one.

Connect your whole account

A link covers one server. To give Claude - or any MCP client - your account instead, connect it once.

In Claude.ai

Add this address as a custom connector:

https://panel.renode.space/api/client/mcp

In Claude Code

Run this command:

claude mcp add --transport http renode https://panel.renode.space/api/client/mcp
  1. 1A browser opens on the panel to approve the connection
  2. 2Choose which servers the assistant may reach
  3. 3Choose whether it may manage files only, or also start and stop them
  4. 4Access refreshes itself; disconnect at any time under Account → Connected AI

Manage this server

Every other purpose lets the assistant read. A "Manage this server" link also lets it act - inside a fixed boundary.

It can

  • Start, stop and restart the server
  • Send console commands
  • Read and edit files

It never can

  • Delete the server
  • Reinstall it
  • Touch databases or backups
  • Change users or settings

Anything outside that list is never possible through a link. Every action is logged on the server's Activity tab, and a Manage link stops after 24 hours unused.

Pre-flight check

On the GitHub tab, the panel inspects the linked repository and shows what will break before a deploy.

  • The start script
  • The port
  • The build step
  • The Node version
  • A missing .env

No AI involved - the panel reads the repository itself.

Context written into the repository

A switch on the GitHub tab opens a pull request that adds .renode/server.md and a block in CLAUDE.md or AGENTS.md, so coding assistants read the server's rules automatically - no link needed.

What the assistant reads

Shortened, with illustrative values. The real one is filled in from your server.

Example
You are helping me run a project on a RE:NODE server.

Task: set up a new project

Server
- Runtime: Node.js 20
- Port: read it from the SERVER_PORT environment variable and listen on 0.0.0.0
- Code lives in /home/container
- On every start the server fetches the branch below, then runs the start command

GitHub
- Repository: your-name/your-app, branch main
- Redeploy on push: on

Databases
- MySQL database s12_app (connection details are in the panel; no password here)

Limits
- 1 GB memory, 10 GB disk

Rules
- .env is not in the repository. Tell me which variables it needs;
  I will upload it through the Files tab.
- Never ask me to paste a password or a token into this chat.

What it tells the assistant

  • The runtime the server runs
  • The port - and that the code must read SERVER_PORT
  • The order the server boots in
  • That the code lives in /home/container
  • The GitHub repository, branch and auto-deploy settings
  • The databases attached, without passwords
  • The server's limits and the rules of the platform

No secrets are shared

Whatever reaches the assistant is written as if somebody else will read it.

  • No passwords, no tokens

    Database passwords, API keys and deploy tokens are never shown to the assistant - not through a link, not through a connected account. It is told that they exist and where they live, not what they are.

  • Your .env stays yours

    The .env file is not in the repository, and the assistant is told so. It tells you which variables it needs; you upload the file yourself through the Files tab.

  • Everything it does is logged

    Every action taken through a link or a connected account appears on the server's Activity tab.

  • Access ends without you

    A link stops by itself once it goes unused. A connected account can be disconnected at any time under Account → Connected AI.

Questions

Which assistants does it work with?

Claude, ChatGPT or any assistant that can open links. The same link followed by /mcp works as an MCP server URL, and Claude or any MCP client can be connected to your whole account.

How long does a link work?

There is no duration to choose. A link keeps working while it is used and stops by itself after 7 days unused (24 hours for "Manage this server"), with a hard limit of 90 days (30 for Manage). A new link for the same purpose replaces the old one.

Can the assistant delete my server?

No. A "Manage this server" link can start, stop and restart the server, send console commands, and read and edit files. Deleting the server, reinstalling it, databases, backups, users and settings are never possible through a link, and every action is logged on the Activity tab.

What is the difference between a link and connecting my account?

A link is for one server and one purpose. Connecting your account gives Claude or another MCP client the servers you approve, with files-only or start/stop access as you choose; access refreshes itself until you disconnect it under Account → Connected AI.

My server won't start. What do I paste?

Choose "Server won't start". Instead of a link it copies a prompt with the latest console lines, so the assistant reads the actual error.

Does the assistant see my passwords?

No. No password or token is ever shown to the assistant. Never paste one into a chat; upload .env yourself through the Files tab.

Try it on a server you already have

Open a server in the panel, pick a purpose on the Console tab and press Connect AI.

Open the panelRead the guides