Repeated Word Proofreader

How to use Repeated Word Proofreader

Review a draft for accidental duplicated words or pasted paragraphs without changing the text automatically. Word runs retain each original spelling and UTF-16 span; paragraph matches preserve their full source evidence. Repetition can be intentional, so the report is a set of review suggestions rather than a grammar verdict.

1. Paste a draft of at most 50,000 well-formed characters.
2. Choose exact word comparison or ASCII A–Z lowercasing; paragraph comparison always remains exact.
3. Review original spans and paragraph evidence before making any editorial change yourself.

Good to know

• At most 50,000 UTF-16 source characters, 10,000 retained word segments, 1,000 nonempty paragraphs and 500 total repeat groups. Empty input returns an empty report; no default text or hidden content is inserted.
• Words use this runtime Intl.Segmenter with English word granularity and retain word-like segments containing a Unicode letter. Only whitespace may separate a repeated run; punctuation, numbers or other intervening source content break adjacency. Runtime segmentation data can differ between browsers.
• Comparison is exact or ASCII A–Z lowercasing only. It does not perform Unicode case folding, accent removal, normalization, stemming, spelling correction or language detection. Reported offsets are zero-based half-open UTF-16 spans in the unchanged source.
• Paragraphs are runs of nonblank physical lines; a blank line contains only spaces or tabs. Boundary spaces/tabs are ignored, while internal spelling, line endings and whitespace remain exact. Intentional phrases such as “that that” may be flagged; no deletion or rewrite is performed.
• Processing is local. Output is limited to 1,000,000 text characters unless a stricter file limit is stated.