[Bug]: CloudFlare Captcha Bypass is Inconsistent using Brave Browser

[EthanC]

Checklist before reporting

• I have searched for similar issues and didn't find a duplicate.
• I have updated to the latest version of pydoll to verify the issue still exists.

pydoll Version

2.8.1

Python Version

3.13.7

Operating System

Windows

Bug Description

I'm unsure whether this issue is a result of using Brave (a Chromium-based browser) or if it also occurs using a standard Chrome installation.

When navigating to the site https://jackbit.co/, the user will consistently encounter a CloudFlare Turnstile Captcha on their first page load of the session. PyDoll will sometimes successfully bypass this Captcha, while other times it will indefinitely hang as if it cannot find the Captcha.

Steps to Reproduce

1. Import the library using import pydoll
2. Construct a Chrome() instance with the binary_location option set to the Brave browser path
3. start() the browser and set_window_maximized()
4. enable_auto_solve_cloudflare_captcha() in the tab
5. Navigate to https://jackbit.co/en

Observe page will often hang, awaiting verification.

Code Example

from pydoll.browser import Chrome
from pydoll.browser.options import ChromiumOptions
from pydoll.browser.tab import Tab

async def main():
    options: ChromiumOptions = ChromiumOptions()

    options.binary_location = env.str("local/path/to/brave.exe")

    browser: Chrome = Chrome(options)

    async with browser:
        tab: Tab = await browser.start()

        await browser.set_window_maximized()
        await tab.enable_auto_solve_cloudflare_captcha()
        await self.tab.go_to("https://jackbit.co/en")

Expected Behavior

PyDoll should bypass the CloudFlare Captcha and continue to the desired page.

Actual Behavior

The PyDoll-driven browser window stays on the CloudFlare Captcha page until the user manually interacts (non-headless) with the window to bypass it.

Relevant Log Output

Additional Context

In my limited time with PyDoll, all other usage of Brave has worked as expected. So, I'm inclined to believe this is not a browser-specific issue.

[thalissonvs]

Hey @EthanC, thanks for reporting! I'm taking a look at this

[thalissonvs]

Can you try this?

    await tab.enable_auto_solve_cloudflare_captcha(time_before_click=5, custom_selector=(By.ID, 'AJXH4'))
    await tab.go_to('https://www.planetminecraft.com/account/sign_in/')
    await asyncio.sleep(10000)

This is an example url, but sometimes you should set a custom selector. This must be the div containing the shadow root. Also, you can set a time before clicking

[EthanC]

Thanks! It looks like setting the selector ID has worked for the time being. But it seems like this ID changes to something randomized after a couple of days. I take it there's no solution that wouldn't require maintenance?