When AI writing feedback quotes a student's sentence back to them, did it copy the student's words exactly, or quietly fix them? Paste the student's text and the feedback; every quoted passage in the feedback is checked against the student's text. Nothing leaves this page: it works offline and makes no network requests.
Similarity is the word-level match ratio (0 to 1) between the quote and the closest passage, the same measure Python's difflib uses. Single quotation marks are not treated as quote marks, because they double as apostrophes; use "one quote per line" for those. This is a mechanical check of copying, not a judgement of the feedback's quality.
Why Markdown-aware: in MV2's audit of 50,742 quotes in published AI feedback on IELTS essays, only 37.7% were character-exact, but 95.7% kept the student's words once formatting and ellipses were set aside. Changed words were most common in short quotes of 4 to 6 words, and some of those are suggested phrasing rather than the student's words, so read short changed quotes first.
Where to look: in MV2's audit of the same IELTS feedback, close copies with a word changed were mostly in short quotes (3.19% of close 4-6-word quotes against 0.74% of close quotes of 11 or more words), and when long quotes changed, two in three were loose matches, more like a reworded passage than a one-word slip. So the page counts the words that differ: 1 or 2 reads as a close copy, 3 or more as a looser match. That cut is a reading aid, not a finding; only reading the quote tells a misquote from a suggested phrase.