Skip to content

Using the app

Browser control

Let the AI read, click and type on websites, only where you allow.

Some sites only show their content after JavaScript runs, or need a few clicks to get to the answer. For those, the AI can use a browser panel beside the chat.

Using it

  • Switch on Browser from the + menu, or just ask: "open my portfolio in the browser and list the projects".
  • The panel opens beside the chat with back, forward, reload and an address bar. You can browse in it yourself too.

Permission per site

The first time the AI wants to act on a site, it asks. Allow lets it use that site until you close the app; Deny stops it. The sites it may use are listed in the panel and can be removed. Reading and scrolling the page that's already open don't ask.

What it can do

  • Read the page, including content built with JavaScript
  • Click links and buttons, type into fields, scroll
  • Play, pause and change the volume of videos

What it won't do

  • Type into password, one-time-code or card fields. It asks you to do that.
  • Open private or local network addresses on its own.

Everything happens in front of you; you can stop the answer at any time.