Unicode inspector

A header line with the number of unique code points, the total character count and how many of those are non-ASCII, then a table. Each table row gives the code point in U+ notation, the character itself, its Unicode category, and how many times it occurs. Rows are sorted by count, most frequent first.

Eight buckets, decided by Unicode property: letter, number, punctuation, symbol, space, control, format and other. Space means a space separator specifically, control means a C0 or C1 control character, and format covers the invisible formatting characters such as the zero-width joiner and the directional marks. Anything that fits none of those, including unassigned code points, shows as other.

Because the character has no visible form. Control and format characters are shown as a middle dot placeholder so the table stays aligned and the row remains readable. The U+ number in the first column is the real identity, so use that to look the character up or to search for it.

That is the cap on the table, and a line underneath tells you how many more unique code points were not shown. The header counts are complete, so you always know the true number of distinct code points even when the table is truncated. Rows are chosen by frequency, so what you lose is the long tail.

This one is an inventory, that one is a check. Unicode inspector lists every distinct code point in the file with its category and count, and does not judge any of them. The detector looks for a specific list of 18 invisible characters and about 35 Cyrillic and Greek homoglyphs, reports each occurrence with its position, and says Clean when it finds none.

You get "Empty input" and a note that there is nothing to inspect, rather than an empty table with a header of zeroes. Everything runs in a Web Worker on your device, and the file is only read, never modified and never returned, so there is nothing to download afterwards. The tool is per-file, so a batch gives you one report per file.