How to pass a JS init script to the BrowserConfig?

[SlitiBrahim]

Hey guys, amazing job made on Crawl4AI, well done.

I have a question. So, when I'm instantiating a Playwright instance, I can pass some init script.
My goal here is to avoid bot detection. And as you can see on the following screenshot, I can pass the webdriver anti-bot detection with this PlayWrite code.

async def setup_stealth_browser():
    """
    Launches the browser with your exact configuration and applies stealth patches.
    Returns: playwright, browser, context, page
    """
    p = await async_playwright().start()
    browser = await p.chromium.launch(headless=False, slow_mo=50)
    context = await browser.new_context(
        viewport={"width": 1280, "height": 800},
        user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36",
        locale="en-US",
        timezone_id="Europe/Paris",
    )

    await context.add_init_script("""
        const canPlayType = HTMLMediaElement.prototype.canPlayType;
        HTMLMediaElement.prototype.canPlayType = function(type) {
            if (type === 'video/mp4; codecs="avc1.42E01E"') {
                return 'probably';
            }
            if (type === 'video/webm; codecs="vp8, vorbis"') {
                return 'probably';
            }
            if (type === 'video/ogg; codecs="theora"') {
                return 'maybe';
            }
            return canPlayType.call(this, type);
        };

        delete Object.getPrototypeOf(navigator).webdriver;
    """)

    page = await context.new_page()
    return p, browser, context, page

But I want to do the same thing with my Crawl4AI browser config instance because without passing this code, as you can see on the following screenshot, the bot may be detected because of the webdriver test.

The code:

    browser_config = BrowserConfig(
        headless=False,
        cookies=cookies,
        user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36",
        viewport={"width": 1280, "height": 800},
    )

    crawler_config = CrawlerRunConfig(
        cache_mode=CacheMode.BYPASS,
        verbose=True,
        # log_console=True,

        wait_for="domcontentloaded",  # More specific selector
        simulate_user=True,
        extraction_strategy=None  # Disable extraction for debugging,

    )

    async with AsyncWebCrawler(config=browser_config) as crawler:
        result = await crawler.arun(url, config=crawler_config)

        if not result.success:
            print("Error on scraping")
            return

        print(result.extracted_content, result.status_code)

        input("Press Enter to close...")

Crawl4AI==0.5.0.post4

Thank you for your answer.

[SlitiBrahim]

any idea?

[SlitiBrahim]

Okay, I realized that I was not doing Crawl4AI the right way and instead I should use something like Playwright to navigate through pages, do all the interactions and not be detected by Antibot system and just use Crawl4AI to extract the data from those pages.