your feed, your filter
SlopRadar uses your browser's built-in Gemini Nano model to detect AI-generated marketing slop and engagement bait on LinkedIn, X, Reddit, and Threads — and quietly hides it. 100% on-device. Nothing leaves your browser.
Spots engagement-bait posts, vague "this image" framing, recycled corporate platitudes, and the other tells of AI-generated marketing content.
Different rules for news, commentary, and other posts. Wire-service breaking news isn't slop just because it's terse and in ALL CAPS.
Confirm or reject any flagged post. Mark your own slop with right-click. The filter sharpens with use.
Verdicts are cached so reloads are instant. Promoted posts are caught structurally before the model even sees them.
The moment your first scan of the day lands, SlopRadar builds a morning brief on-device: how much slop it filtered, the day's biggest stories across your platforms, trending tickers, and a two-sentence AI summary.
Follow a developing story across LinkedIn, X, Reddit, and Threads with one click — the query builds itself from the post. Each story extracts its central claim, and incoming posts vote: ✅ confirms, ❌ contradicts. Watch a rumor hold up or fall apart.
Echo chains map the same story's appearance across feeds in arrival order — organic stories spread messy, campaigns move in lockstep. Flag a push, and SlopRadar finds the network behind it and can hide it wholesale.
Every account you've scanned carries a credibility dot — green for a clean history, red for serial sloppers — computed entirely from your own capture history. No external ratings, no server.
An experimental research bench: backtest social-sentiment trading strategies against a real comment corpus, fork strategy variants, and track out-of-sample performance — all computed locally.
Invite five friends who stick around a week and get a month of Pro free. Referral tracking is a single tiny check-in for the referee's first week — and nothing, ever, otherwise.
Add the extension from the Chrome Web Store. Visit LinkedIn, X, Reddit, or Threads. SlopRadar starts watching as you scroll.
Each post's text is sent to Chrome's on-device Gemini Nano model. The model returns a verdict in a few seconds — entirely inside your browser.
Slop posts get a clear red banner with the reasons it was flagged. Click "Show anyway" to reveal, or "Not slop" to teach the filter.
All classification runs locally using Chrome's built-in Gemini Nano model. There are no servers, no analytics, no tracking, and no third parties involved. Your feed stays yours.
Reddit and Threads support is in beta — selector coverage may shift as those platforms iterate.
SlopRadar is free, open about how it works, and runs entirely on your device. Install it and your feed gets quieter in a few minutes.