Invisible Character Detector

In short When a value looks identical to the one you are comparing it against but the lookup still fails, the answer is very often a character you cannot see. Zero-width spaces from a copy out of a web page, a byte order mark left by a Windows editor, non-breaking spaces pasted from Word, or a directional override sitting between two visible letters. Twenty categories are detected here, including U+200B, U+FEFF, U+00A0 and U+202E, and the highlighted view shows exactly where each one sits so you can decide whether it belongs before deleting it. Some of them are legitimate: non-breaking spaces exist for a reason and zero-width joiners hold emoji sequences together. The one that is never innocent in ordinary text is U+202E, the right-to-left override, which reverses everything after it on screen while leaving the underlying bytes untouched — that is how a file called "exe.txt" can display as "txt.exe" and vice versa. These characters come from copying out of rendered documents rather than from typing, which is why they appear in pasted config values, spreadsheet exports and anything that has been through a word processor on its way to a code editor.