Media
2 items
Description
Your yarnnn agent works in Chrome for you — in its own tabs, with your sign-ins — doing what you ask, as you. yarnnn for Chrome lets your yarnnn agent carry out web tasks in your own Chrome when you ask for them in yarnnn: open a page, read it, click, fill in a field, drag, type, go back. Because it runs in your browser, it can work on sites you're already signed in to. You need a yarnnn account (www.yarnnn.com). The agent works only in its own tabs, kept in a tab group named "yarnnn" that you can see and close at any time. Your own tabs are never touched. It keeps working if you close yarnnn, as long as Chrome is open. When a page draws what the agent needs instead of labelling it — a calendar's grid, a board's cards, a canvas — the extension can point, drag and type in its own tab and look at a picture of that tab, through Chrome's debugger. Chrome shows a bar saying yarnnn is debugging the browser while it does; closing the bar stops it in that tab. A picture is used for that one step and never kept. Before the agent acts on a site for the first time, the extension asks you in its own window. Your answer is kept for that site. From the toolbar popup you can switch the extension off and review or remove the sites you've allowed or denied. Some sites are never allowed, whatever you answer: banking and payments, trading and crypto, password managers, and account security pages. A password field's value never leaves the page. The extension connects to your yarnnn account once you open yarnnn in this Chrome, and receives the agent's steps only over that connection. A yarnnn page can switch it off, but can never hand it a step. All of its code ships in the package, and it runs no remote code. Available in English and Korean.