Slop Mop is a free, open-source Chrome extension that helps reduce AI slop and low-value writing in LinkedIn feeds. It reads posts ahead, explains what it noticed, and lets you hide, highlight, reveal or override. Showcase project of how to use Jev as a decision model along with tells of bad writing (9 evaluations and 2 usefulness factors). It is a research project not a product so it's free to use and fork!
Hey Product Hunt! I'm Tom. I built Slop Mop because my LinkedIn feed filled up with engagement bait and "I turned down a 7-figure offer at 24" stories, plus a lot of furthermore-flavored wisdom (Graphite's research clocked "furthermore the" at 43x the human rate, and my feed agrees).
Slop Mop is a free, open-source Chrome extension for desktop LinkedIn. It reads posts before you see them and folds suspected slop into a strip you can reopen, or outlines it and shows its work. You set the threshold, and you can override any call.
The idea I'm most excited about: it's not an AI detector, it's a bad-writing detector. Eleven checks from Jev (a decision-probability model by Typesafe AI) judge each post's writing, and useful posts survive no matter how they were written. And Jev's inference is cheap enough to judge every post as you scroll, continuously, for free. That used to be too expensive at feed scale. It isn't anymore.
Free forever, MIT licensed, no account, nothing about you nor the author is processed or evaluated. It's live on the Chrome Web Store now.
I'd love your feedback, especially the embarrassing examples where the mop gets it wrong.