Skip to main content
Use the built-in OpenWork Browser for websites. To work in a native Mac app, use the separate Computer Use extension. OpenWork can open pages, click, fill forms, read page content, and take screenshots inside that browser. If you connect to a remote OpenWork server, the browser runs on that remote worker. Today, the supported path is the first-party OpenWork Browser extension.

​
Requirements

  • Use the desktop app or a remote worker that exposes the OpenWork Browser.
  • Keep OpenWork Browser enabled in Settings > Library.
Then go back to the session view and ask OpenWork to open a website.

​
Allow control for a thread

The browser panel asks Allow browser control for this thread? Choose Allow for this thread once to let that conversation navigate allowed websites, read pages, and scroll across its own tabs. You do not need to approve every page or scroll. Organization restrictions still apply. Clicks, typing, and keyboard input still ask Allow browser action? before they run, because they can submit information or change website data using your signed-in account. Review the target and text, then choose Allow once or Deny. Time spent reviewing does not use up the short page-observation window, but a changed page or target still requires a fresh observation. Browser operations remain time-limited. Keyboard automation requires a directly focused native field, button, or link; embedded and custom editors with unverifiable focus require you to take over. Choose Take over at any time to pause browser operations and revoke the grant. Sign in directly in the page, then choose Resume browser; the next browser operation asks for a fresh grant. Disabling browser control, canceling an in-flight browser operation, closing the thread’s last browser tab, or restarting the desktop also clears it. The grant never transfers to another thread. Website tools (WebMCP) still require separate approval to run, and their results stay local until you choose Share result. Browser control does not grant access to your regular Chrome profile or other conversations’ tabs. In the desktop browser toolbar, Suspend frees a tab’s memory after a warning. This is manual, not a page snapshot: form input, scroll position, and page history are lost. Reload or selecting the suspended tab opens its saved URL again. Tabs opened or restored for browser work stay protected until browser.release_tab declares all running and queued work complete. Release does not suspend or invalidate the current target; there is no automatic reclamation. For later work, browser.restore_tab checks conversation ownership and returns a fresh protected handle to use.

​
Keyboard shortcuts

In the desktop app, these shortcuts apply when a built-in browser page, its URL field, or its browser controls have focus: The native app menu’s Close command also closes the focused browser tab before closing the window. Reopening creates a fresh page from a URL kept only in memory, not a saved page snapshot. It does not restore form input, scroll position, or page history, and closed-tab URLs do not survive an app restart. After closing the last browser tab, press the reopen shortcut immediately. Clicking elsewhere, pressing Tab, or changing conversations clears that immediate browser context. With the chat composer focused, Cmd+T / Ctrl+T still cycles conversations instead of reopening a browser tab.