Will it connect to my server?
It will connect if you grant access. On the cloud plan you cannot grant it however much you want: command execution there is off hard. Put it on your own server — you switch commands on yourself, with a toggle in your account, and there you also name the ones that must ask your confirmation.
What if I have no documentation at all?
You usually do have some, it’s just scattered: notes in a text file, comments in configs, messages to yourself. Feed it whatever you find — from there it answers from that, and you add a line right in the chat whenever you change something. What you never gave it, it doesn’t invent.
Can it read my configs?
Yes, as ordinary text files: docker-compose, web server configs, systemd units, your markdown notes. It finds the right place in them and explains it in words, naming where it took that from. It will not apply or edit them: you change the files yourself.
Where do we write to it?
Right where you already are: Telegram, WhatsApp, VK and MAX. Every conversation lands in one inbox, and there’s a mobile app too — from somebody else’s laptop that is sometimes the only way to recall your own setup.
What is the file it sends?
A table — CSV or XLSX — that it builds itself and sends as an attachment into the chat. That’s how service and port inventories, volume lists, backup schedules and domain lists go out. The file opens right where you asked.
Does it remember what I’ve already fixed?
Yes. It pulls up your earlier messages and answers with them in mind: what broke in the spring, how it ended, what you decided not to touch. The limit is an honest one — if you fixed it in silence and it never reached the chat, it doesn’t know.
Can I run it on my own server?
Yes, and for this audience that is usually the right answer. The same product installs on your own machine, and conversations never leave it. The free Community plan gives you one agent and Telegram, forever and without a card; the cloud is for people who don’t want to run yet another service.
What has to be done before the first answer?
Connecting a model is the first step in the dashboard, and you do it once. Until a model is connected the assistant stays silent: the thinking happens in the model, not on our server. The model can be a local one, on your own hardware.
Do I need a separate assistant for each setup?
This plan includes one assistant, and for a home server that’s usually enough: notes on the network, the media library and your experiments all sit together. If you want to split setups apart, there is room for that on the next cloud tier — the pricing page says exactly how much.
Can I see it before paying?
Yes. The demo sandbox is open without sign-up — login and password below, it resets every hour, so break whatever you like.
Fits into what you already use every day
Telegram
WhatsApp
Don't take our word for it — check for yourself