Blah blah!

I built a LinkedIn cringe detector and it immediately flagged my own posts. That's not a bug. It's the whole point. LinkedIn has its own dialect and it gets in. You don't notice you're absorbing it because everyone around you is doing it, and after enough years on the platform you wake up writing things like "I'm passionate about delivering impactful solutions" and you don't even flinch. The phrases that used to sound like obvious LinkedIn cosplay just… sound like business now. This extension highlights it. Not just AI-generated text, although yes, that's a category, but the whole family of recognisably hollow LinkedIn moods. It looks for six kinds of cringe and shows you what it found: 🫠 AI corporate slop: humbled-to-announces, in-today's-fast-paced-worlds, tapestries, "Thoughts? 👇", the LLM vocabulary set (delve, leverage, multifaceted, harness, elevate) ☠️ Doomer: "AI will take your job", "we're cooked", "junior devs are dead", the rolling apocalypse forecasts 🤖 Vibe coder bro: "$50k MRR in 7 hours with Cursor", "no engineers, no funding", "AI is my cofounder". Gets a parental advisory sticker, naturally. 💪 Hustle culture: 5am clubs, "while you were sleeping I was…", "sleep is for the weak", the whole grindset cinematic universe 🎣 Engagement bait: "drop a comment", "reposts appreciated", "tag someone who", "free guide", "let that sink in" 📐 LinkedIn-shaped: one-line paragraphs, arrow bullets, hashtag piles, the caps opener. The formatting the algorithm taught everyone, scored on its own so you can switch it off Flagged posts get a yellow-to-red heat-map overlay (gentle yellow for a bit LinkedIn, glowing red for pure slop) and a 1–5 cringe score in the corner. The original text stays put. You read it. You judge for yourself. A pill in the bottom-left corner lets you dial the sensitivity per category: off, lenient, normal, ruthless. Want the doomers gone? Set doomer to off. Think the slop detector is being too generous? Crank it to ruthless and watch your feed catch fire. Hover a label and it tells you exactly which phrases scored, the worst one, and what to say instead. Start typing a post and it scores you as you go, before anyone else sees it. Someone's own news, a new job, a job hunt, is never flagged: the mirror is pointed at the language, not the person. The honest small print: this is heuristic, not a real AI detector. No reliable AI detector exists. The extension catches phrases that read like AI slop, whoever wrote them. Don't use it to accuse anyone of anything. Use it to see how much of your feed reads like the same five LinkedIn posts in a trench coat. It runs on linkedin.com only, entirely in your browser. No analytics, no telemetry, no remote server. The only things it stores, on your machine, are your settings and a count of the posts you rescued this month.