TF

tried

• nodriver has no way of downloading files with some given name. we need to find a way to extract the nodriver session, and then download files using a requests library.
• hrequests doesn't work. BrowserSession() gives weird errors about addons and context, even when using the basic example code from docs.
• playwright gets cock-blocked by google login. it doesn't work, atleast not out of the box.

try next

• patchwright (based on playwright)
• camoufox (based on playwright, hrequests uses it)
• export session cookies from browser, and use them with some requests library. this method would not require browser automation.