Computer use means an agent operates a screen the way a person would: clicking, typing, reading, and navigating. It works through the same interface you use rather than a clean backend connection, which makes it flexible and also clunky. It earns its keep when there’s no API, no connector, or the task depends on what a user actually sees, like visiting a client portal or downloading a report.
Picture asking an assistant to sit at your laptop and handle a website task. They open the browser, log in, read the page, click the button, fill the form, and tell you what happened. They use the same screen you would and never touch the site’s technical plumbing.
Why you care
Computer use is a practical fallback, so reach for a cleaner, safer connection first when one exists. The front door brings friction: the agent may need authentication, hit a two-factor prompt, or trip when a page changes, and permissions matter because an agent using your logged-in browser may reach whatever that browser can. Sometimes the only interface available is the same screen you use, and this is how an agent works with it.