🧽 AI Slop Checker

What gets flagged

Overused words such as "delve", "leverage", "robust", "seamless" and "landscape". None is wrong on its own; they become a problem when they replace a precise word. The list starts from the same words our own publishing system removes from every article before it goes live. Empty phrases such as "in today's fast-paced world", "it's important to note that" or "let's dive in" take up a sentence and add nothing - cutting them almost always improves the text.

How the score is built

The score adds up three parts, and the cards above show each one so nothing is hidden. Filler density counts for up to 60 points: three flagged words or phrases per 100 words gives the full 60. Missing specifics counts for up to 25: in texts of 80 words or more, fewer than one meaningful number (an amount, a percentage, a year, a size) per 100 words adds points. Flat rhythm counts for up to 15: when five or more sentences are all about the same length, the text reads like a list. For comparison, the articles on this site score between 0 and 38, with a median of 17.

What it cannot tell you

This is not an AI detector, and no reliable one exists - people write "leverage synergies" too, and a good AI-assisted draft can score 0. A low score does not make a text good, either: it can be clean and still wrong, dull or unsourced. Use it the way an editor uses a red pen on a first pass, then read the text aloud. Everything runs in your browser; nothing you paste is stored or sent.