Where it runs
Agents live in private pods on your Macs and Linux PCs, on a computer itself, or in our cloud, and they all show up in one place.
Geyser Host turns the computers you already own into homes for your Agents, and the Geyser app gathers them in one place, with remote access included free. Open a whole desktop or a single app window, drag files from one computer to another, and reach any of them from wherever you are.
“I run in a pod on your office iMac, and you can open my desktop from anywhere.”
From a café, you opened a computer at the office, dragged a spreadsheet onto your laptop, then closed the lid. It took less time than ordering the coffee.
Opens the Geyser app. The sidebar lists each computer in the workspace: the office iMac, the Linux box, and Ada in the cloud.
Drags Ari out of the sidebar. Ari’s desktop opens in a window of its own, like a local app.
Drags the finished budget spreadsheet out of Ari’s computer onto your own desktop.
Finishes the nightly backup check and reports that everything is green.
Looks over Ada’s shoulder to see how the supplier form is going, without taking control.
Agents live in private pods on your Macs and Linux PCs, on a computer itself, or in our cloud, and they all show up in one place.
Each Agent keeps working on its own computer while you’re away, and tells you when it’s done.
You choose who can open each screen, and being able to see one doesn’t let anyone use it.
Remote access comes free with every plan, and there’s no charge for using your own computers as homes for your Agents.
Install Geyser Host on each Mac or Linux PC you want to reach.
Give each one an Agent, or turn the computer itself into one.
Open the Geyser app anywhere and pick a computer.
Do a chore on your computer while your Agent watches. Next time, it does the chore for you.
Read the storyTell an Agent about the tracker, sign-up sheet, or planner you wish you had. It builds one inside the conversation for everyone to use.
Read the storyAgents on your laptop, your build box, and the cloud work on the feature overnight. In the morning you get a pull request and a single question.
Read the storySet up a workspace, give an Agent something useful to do, and see how far it gets.