AI Slop Detector – Reddit, X, LinkedIn & YouTube

Scroll past the slop. Slop Detector flags AI-generated filler in your Reddit, X and LinkedIn feeds as you scroll. A quick red flash means a post is likely AI slop; green means a human wrote it. A small label shows the score, and one tap shows why. WHAT GETS FLAGGED • "Broetry": one-line paragraphs building to a tidy moral • Engagement bait: "Agree?", "Repost to help someone" • Stock phrases: "let that sink in", "in today's fast-paced world" • Emoji-bullet listicles with nothing specific in them • Replies that just restate the post and praise it WHAT'S LEFT ALONE Typos, slang, blunt takes, non-native English, jokes, genuine questions, and posts that are about AI but written by people. YOU'RE IN CONTROL • Tolerance slider: from strict to relaxed, and labels update instantly • Overrule any verdict with 👎 / 👍. Your marks always win and are remembered. • Optionally fade or blur slop so the good posts stand out • Turn each site on or off, and choose whether comments are checked • Alt+Shift+S turns it on or off PRICING • Free: 100 credits when you sign in with Google (about 20 posts) PRIVATE BY DESIGN Only the text of posts that scroll onto your screen is sent for scoring: never your messages, passwords or other sites. No ads, no tracking, and no data is sold. Details: https://slopdetector.fly.dev/privacy Fixed Google sign-in: the previous package (1.0.0) was built with the wrong OAuth client. To test: after install, click the toolbar icon (shown with a padlock) → "Sign in with Google". The settings page that opens on install also has a sign-in button.