Your own cloud agent, in minutes.

A free, open-source AI agent that runs on a computer of its own, remembers what you tell it, and talks to you from the chat apps you already use. It can browse, write, keep a schedule, and get things done while you’re away. Everpod gives it a home.

A real computer in the cloud, and your agent’s alone. It’s the Mac Mini people buy to give their agent somewhere to live, without the Mac Mini: it has its own files and its own browser, it installs the tools it needs, and it keeps all of that through every restart and update, along with its memory and the accounts you give it. Nothing on it is shared with other customers, and it stays awake around the clock.

Your calendar and what needs you, every morning. The inbox sorted overnight. A price watched until it drops. The school calendar kept in the family chat. What you told it months ago, remembered. Real people show what they do with theirs, in their own posts: What a cloud agent can do.

No. There’s nothing to install and nothing to code. You name your agent and pay, and we build the computer, install your agent and secure it. Connecting your first chat app is a walkthrough, step by step. From there you talk to your agent in plain language, and the technical work is its job, on its own computer.

An email address and a way to pay: a credit or debit card, Apple Pay or Google Pay. Prices are in US dollars, and the checkout shows your total in your own currency where supported, with tax added where it applies. Everything else is part of the plan: the pod, your agent’s setup, security, backups, and its model usage. No API keys to hunt down, no server to look after.

The pod is our job: we build it, keep it secure and backed up, update your agent’s software, and cover its model usage from your plan. Yours is choosing where you talk to your agent. We walk you through the first chat app, and your agent can set up others when you ask. If you get stuck, email [email protected].

Yes, the way you’d move to a new laptop. A pod is an ordinary computer, so your agent can pull in your existing workspace and memory from wherever they live today, a git repo or an archive, and carry on from there. There’s no one-click import yet: tell your agent where the old one lives, and it does the moving.

OpenClaw today, with Hermes, Claude Code and Codex the ones we’re bringing up next. The pod doesn’t change with the agent: whichever you choose, it gets the same private computer, secured, backed up and kept up to date for you. Want one we don’t host yet? Tell us at [email protected].

No. What you say to your agent, what it remembers, and the files it makes live on your pod, not in our systems: we don’t collect it, read it, or train anything on it. We touch the machine only to keep it running, and we open what’s on it only if you ask us to during support, or if security or the law requires it. Your agent’s thinking goes to an AI model provider, as with any agent; we send included usage only to providers that don’t train on it. Pods are hosted in Europe. The full picture, in plain words: Privacy policy.

You can leave anytime, and your pod runs to the end of the time you’ve paid for. 7 days after that, we delete it permanently: the machine, your agent’s memory, and the backups all go with it. The short grace window is there in case you cancel by accident; past it, we don’t hold on to your data. Want to take your agent with you? Email us before then and we’ll send you a full copy of your pod’s data.