Detect Zero-Width Characters in Text
Zero-width characters are invisible Unicode code points that hide between visible characters, breaking string comparisons, YAML keys, and code identifiers without any visible sign.
The three most commonly encountered zero-width characters are Zero-Width Space (U+200B), Zero-Width Non-Joiner (U+200C), and Zero-Width Joiner (U+200D). A fourth, the Zero-Width No-Break Space / BOM (U+FEFF), appears at the start of UTF-8 files saved by some editors and is invisible when pasted into most tools.
Common sources of zero-width characters include text copied from web pages, PDF exports, Microsoft Word documents, and content generated by AI language models. They pass most linters silently and cause mysterious failures — a YAML key that "looks identical" but doesn't match, a Python variable that shadows another without any syntax error, or a URL that 404s because the path segment contains a hidden character.
The Invisible Unicode Character Detector scans every Unicode code point in your text and reports zero-width characters with their exact codepoints, names, and occurrence counts. The cleaned output strips all detected invisible characters while preserving every visible character.
How to detect zero-width characters
- Open the Invisible Unicode Character Detector and paste the suspect text into the input area.
- Click Analyze. The tool reports any zero-width characters found — including codepoint, name, and count. Zero-width joiners and non-joiners are rated Medium risk; zero-width spaces are also Medium.
- Copy the Cleaned text output and replace the original — all zero-width characters are stripped while your visible content is unchanged.
Related invisible character types
- Full invisible character scan — detects all invisible character categories in one pass
- BiDi override attack checker — targets Trojan Source CVE-2021-42574
- Remove invisible characters — focused on producing clean output
- Hidden character finder — broader invisible-character sweep