Bypass Login Captchas with Selenium: Cookie and Browser Profile Tricks

Bypass Login Methods

When using Python for web scraping or automated testing with Selenium, login captchas—especially text and image captchas with interference lines—are a common obstacle. Two main approaches can bypass them: capture and reuse the site’s cookies after a manual login, or launch the browser with a full user profile that already contains the necessary cookies and settings.

Add Cookie to Bypass Login

For example, after logging into a Baidu account, open the browser’s developer tools (F12), locate the www.baidu.com request, and copy the relevant cookie value highlighted in the screenshot.

Include this cookie in subsequent HTTP requests to access Baidu as an already‑logged‑in user.

Download Browser Driver

Selenium requires a driver that matches the browser version. Download the ChromeDriver and GeckoDriver and place them in the Python installation directory.

http://chromedriver.storage.googleapis.com/index.html

https://github.com/mozilla/geckodriver/releases/

Launch Chrome Browser to Bypass Login

Find Chrome’s profile directory by navigating to chrome://version/, remove the trailing \Default, and prepend --user-data-dir= to create the launch argument.

profile_directory = r'--user-data-dir=C:\Users\xxx\AppData\Local\Google\Chrome\User Data'

Close all running Chrome instances before executing the script; otherwise, an error will occur. The full launch code (omitted for brevity) starts Chrome with the specified profile, preserving bookmarks and logged‑in accounts.

Launch Firefox Browser to Bypass Login

Locate the Firefox profile folder via Help → Troubleshooting Information → Profile Folder , copy the path, and set it in the script.

profile_path = r'C:\Users\guixianyang\AppData\Roaming\Mozilla\Firefox\Profiles\dvm6wqam.default'

Running Selenium with this profile launches Firefox with the previously saved bookmarks and logged‑in Baidu account.

Bypass Graphic Captcha Sites

The same technique works for other sites, such as Jianshu. After logging in once and capturing the cookie, replace the target URL in the script with the Jianshu profile URL; the automated browser will retain the logged‑in state.

https://www.jianshu.com/u/52353ffa8b86

With the login gate opened, you can proceed with web crawling, automated testing, or other tasks. Note that this method may not work on every website, but it is effective for many.