Complete AI Training

MCP server · Developer tools

Heliograph MCP server

by dbhq-uk

Run commands on machines you cannot log into, and get back a timestamped log of what happened.

Flow diagram: you ask your AI “Check disk space on sql01 and show me the log”, the Heliograph MCP server connects it to Far machine, and you get back A log with times on every line.

Heliograph is a free tool that runs commands on a machine you cannot log into yourself, like an air-gapped server or a client's locked-down system. It works by leaving your request somewhere both you and that machine can reach, and picking up the log afterwards. It is handy if you keep asking someone else to run things and paste the output back to you.

What is an MCP server? The 30-second version

On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill, in this case the ability to work with Heliograph. Once it is connected, your AI can publish a command request to a remote machine and read back the log of what ran. You still decide what to ask for; the helper just carries it across the gap for you.

What this MCP server does

You tell your AI what you want to check or run on the far machine. The AI uses this helper to publish that request into a shared place, like a git repository or a file share. On the other side, a small script that someone there already started picks up the request and runs it. The output comes back as a log with every line stamped in UTC, so you can see what ran and how long each step took. Your AI then reads that log and tells you what it found.

Flow diagram: you ask your AI “Check disk space on sql01 and show me the log”, the Heliograph MCP server connects it to Far machine, and you get back A log with times on every line. Click to zoom

What you can do with it

  • Send a command to a machine you cannot log into
  • Follow a request while it runs and see its status
  • Read the full log of the last run, with timestamps
  • Spot long pauses in a log to see which step was slow
  • Check whether the setup will work from where you are
  • Plant the small script the far side needs to run

Try asking your AI

  • “Send a disk space check to sql01 and sql02, then show me the log”
  • “Look at the last run log and tell me where it stalled”
  • “Check whether heliograph is set up correctly on this machine”
  • “Publish a read-only request that lists the running services on the payments host”

What it gives back to you

You get back a log file with one line per output line, each stamped with the time it was produced. Your AI can summarise it, point out errors, or show you the raw text in the chat. If you ask for gaps, it also tells you which steps took the longest and how long each pause was. Nothing on the far machine changes unless the request was marked as an action and the operator allowed actions.

Before you start

What you need

  • The heliograph program installed on your machine
  • A shared place both sides can reach, like a git repository or a file share
  • Someone on the far side to run the small start script once
  • The Claude desktop app or another AI tool that supports MCP servers

Good to know

It can run commands on a remote machine, so make sure requests are marked read-only unless you really intend to change something, and only use it where the operator has agreed.

Install it with your AI

Add Heliograph MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check Heliograph MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.

Sign in Become a member

Who it's for

People who need to check or run things on locked-down, air-gapped, or client-owned machines they cannot log into themselves, such as support engineers, SREs, and consultants.