enable manual browsing to solve captchas

[ari7cr]

Is there a way to take over the browser opened by web-ui and log-in to a website or solve a captcha, and then let the agent continue the task? I'm failing to access websites with captchas, and would like to just click it myself - or even open chrome, visit the website, log in and solve captcha, and only then start the automation.. Is that possible with web-ui?

[vvincent1234]

you can pause the agent

[ari7cr]

True, thx. Does visiting google gemini work for you via Chrome or Chromium? In both I only get a blank screen, while other websites work normally. Chrome is using the binary file path, and chromium via the patchright install.

[gitwittidbit]

you can pause the agent

and then? Where/how can I access the browser?

(If that sounds like a strange question: I am logging in from another machine on the network. So, all I have is the gradio webui.)

[vvincent1234]

you can pause the agent

and then? Where/how can I access the browser?

(If that sounds like a strange question: I am logging in from another machine on the network. So, all I have is the gradio webui.)

use vnc