Been thinking about that devtools trick again after seeing someone pull Google's site_quality_score and rq_semantic_class straight from the traffic. It's wild how much is sitting right there in plain sight, like those ChatGPT source sets or even just seeing Reddit being used behind the scenes despite the lack of citations. I've tried poking around the network tab before but never really looked for
actual parameter values like that. Honestly feels like we're all just guessing when we aren't looking at the raw requests.
full read:
https://www.searchenginejournal.com/wiretapping-the-serp-reverse-engineering-network-data/593419/