What PlainSpeak cannot do

PlainSpeak is a finishing tool. This page lists what it does not do, so you know when to rely on it and when to use your own judgement.

The scan

It reads wording and a few patterns, not meaning. It matches stock phrases and a handful of shapes, such as sentences all the same length or each clause picking up the last one's word. A fluent machine draft that avoids them can score clean.
It is not an AI detector. A clean score says a reader is unlikely to trip over the wording. It makes no claim about what a detection service would say, and it cannot prove who or what wrote a piece.
Style is described, not scored. Passive voice, hedging, long sentences and the rest of How it reads are shown but never change the score, because a contract, a paper and a social post are all right to differ.
Short texts give weak numbers. Under about 120 words the voice number after a rewrite is marked low-confidence.

Models vary. Some shorten text when asked not to, some add em dashes, some keep the habits they were asked to drop. PlainSpeak catches the worst cases and keeps your original, but the fix is often another model.
It needs your own AI key and a licence. Scanning is free; rewriting is not.
It cannot remove a hidden watermark that a model's maker may embed in its choice of words, other than by rebuilding the text. Recompose with a model from another family is the strongest option for that.

It is mechanical. It compares numbers, dates, links and quotations between your original and the rewrite. It cannot tell you whether a claim still means what it meant.
Find sources needs an OpenRouter key, and only marks a source as checked when the page really contains the claim's figures.

It needs OpenRouter, and runs on Rewrite and Recompose only.
It is two more opinions, not proof. Models disagree, and after a few rounds what is left is often a matter of taste.