The Browser Agent doesn’t seem to work at all. Sometimes the Browser tool says its “Ready (Chrome detected)” but more often it just says “Unable to check status”. In either case the agent does not work.
Steps to Reproduce
I have it enabled in the settings and I’m in Agent mode. I prompt something like “navigate to google.com” and it just hangs.
Expected Behavior
The browser agent should be able to use the browser in any way.
Hey, thanks for the report. This is a known issue and we’re already looking into it.
Quick question: Are you using Agent mode in the standard editor window, or are you clicking “Open in Agent Window”? Several users noted that Browser tools work in the main editor but not in the separate Agent Window.
Could you try running the Browser Agent in the standard editor (not in the Agent Window) and let me know if that makes a difference?
I just arrived here investigating the same issue and can confirm that I was getting it in agent mode but not in standard. When I check the settings and Tools & MCP in the standard mode I can see Browser Automation Ready(Chrome detected) but when I tried the same setting in the agent mode I saw “unable to check status” rather than “Ready (Chrome Detected)” Switching back to the standard editor I’m able to use the browser automation again. Hope that helps narrow down the issue a bit. I’m quite new to the product but I believe I was opening an agent window.
