Universal AI Slop Finder
By pizzaman · 2 months ago · 5k+ installs
Detects likely AI-written comments and replies on X, Reddit, and YouTube using heuristics inspired by common AI writing patterns. Highlights suspicious comments with scored badges, inline visual marking, sensitivity controls, and live rescanning for dynamic feeds and comment threads. Useful for spotting ChatGPT-style writing, LLM slop, overly polished bot comments, synthetic engagement, and other machine-generated social content.
Key Highlights
- Adds scored AI suspicion badges to flagged comments and replies
- Highlights suspicious text inline with visible severity levels
- Shows tooltips with scores and reasons for each flag
- Rescans dynamic feeds and comment threads as new content appears
- Can dim stronger matches without hiding or removing them
Details
Universal AI Slop Finder marks comments and replies on X, Reddit, and YouTube that look likely to be AI-written. Flagged items get visible badges with suspicion levels and scores, colored highlighting, and hover details explaining which signals triggered the label.
The tweek keeps scanning dynamic feeds and comment threads as new content loads, so fresh replies and posts can be marked automatically. It can also dim stronger matches, but it does not remove, filter, reorder, or confirm that any comment is truly AI-generated.
Runs on
4 sites
Developer
Tweek Code
989 lines ·