for founders and two-person product teams
Your AI product needs an address, not a Docker tutorial.
AgentHotel turns one VPS into a hotel for agents. Each guest gets its own container, its own domain with automatic HTTPS, and memory that outlives the container.
The front desk does the boring parts
Check in
Pick a runtime, name the agent, done. Domain, TLS and routing are automatic, and Caddy issues and renews a Let's Encrypt certificate for every agent subdomain.
Guests remember
Sessions, memory and workspace files live in volumes that outlive the container — so a redeploy does not wipe the thing your product depends on.
Room keys
Provider API keys are configured once and injected into every agent that needs them. A new guest is productive on its first message.
No vacancy
The panel warns you before you overbook the host, not after, and each agent is capped at its own RAM and CPU so no guest starves the others.
Housekeeping
Stopped containers, dangling images and build cache are reclaimed on a schedule, with history and space-reclaimed stats.
The dining room
Guests share a data store, so an agent can find what a colleague already worked out instead of starting from nothing.
What you stop doing
Running AI agents on your own server otherwise means hand-rolled Docker commands, manual reverse-proxy config, and API keys copy-pasted into env files. AgentHotel removes all of that — the panel owns the routes, the certificates and the key injection.
On a fresh Ubuntu or Debian VPS, as root, the whole install is:
git clone https://github.com/magnusfroste/agenthotel.git
cd agenthotel && ./install.sh
install.sh installs Docker, Git and lsof, creates a 2 GB swapfile if the host has none, builds the images with the checked-out commit baked in, and starts everything. No separate docker compose up afterwards, and no parameters. It prints a setup link with a one-time code, so a stranger cannot claim a fresh panel.
Each guest gets its own room — an isolated container with its own memory and CPU allowance, its own domain with automatic HTTPS, and persistent state that survives restarts, redeploys and nights.
AgentHotel README · on why the metaphor is not decoration
What founders ask first
Do I need to know Docker?
No. The installer handles Docker, Git, the swapfile and the first start. You open the link it prints, create your admin account, set your panel domain and deploy your first agent.
How big a server do I need?
The panel itself is light — 1 vCPU and 1 GB runs it fine. 2 vCPU / 4 GB covers the panel plus one or two agents, and each Hermes or OpenClaw agent typically needs about 0.5–1 GB RAM of its own.
What happens on a redeploy?
Memory, sessions and workspace files are volumes, not container state. They outlive restarts, redeploys and nights. Keys added to a provider after the agent existed are injected on the next redeploy.
Which model does my agent run?
Whichever you point it at — a frontier model through a built-in provider, your own model on your own hardware, or a shared endpoint. Switching model or provider is a live choice in the agent's model picker, not a redeploy.
Can I leave later?
Yes. No lock-in: one hotel issues a one-time link and the other fetches the agent from it, server to server, memory included. The agent stops in the old hotel the moment it is pulled.
Give your agent a room tonight.
A fresh VPS, one command, and a guest at its own address with a certificate that renews itself.