This page does two jobs with characters you cannot see. The first part gives you copy-ready invisible and blank characters, such as the Hangul filler (U+3164), the Braille blank (U+2800) and the zero-width space (U+200B), each with its code point and a short note on what it is used for.
The second part is a hidden character detector and text cleaner. Paste text and it highlights zero-width characters, odd spaces, soft hyphens, byte order marks, direction controls and similar Unicode characters, then gives you a cleaned copy with only what you chose to remove.
It is a plain text cleaner for tidying copied text, fixing code and data, and understanding why something looks fine but behaves wrongly. Everything runs in your browser, and your text is never uploaded.
Reviewed by: CalcMojo Editorial Team
Copy an invisible or blank character
Pick a character from the list and press Copy. You can copy several at once with the "How many" box. Then paste it where you need it, for example into a profile name field, a text message, a spreadsheet cell or a test case.
Which one to choose depends on where you paste it, because every site decides for itself what it accepts.
- U+3164 Hangul filler is a letter-like character that most fonts draw as empty space. It is often tried first for a blank-looking name.
- U+2800 Braille pattern blank is a symbol with no dots. It has width, so it behaves like a visible blank.
- U+200B zero-width space has no width at all. Because it is invisible by definition, many forms and chat apps remove it.
- U+00A0 no-break space looks like a normal space but does not wrap, and some fields that reject a plain space accept it.
We have not tested these on every app or website, and sites change their filters. A character that works on one platform may be rejected or stripped on another. If the first one does not work, try another from the list.
Why invisible characters exist
Invisible characters are part of Unicode, the standard that defines text for almost every language. They have real jobs:
- Zero-width space (U+200B) marks a place where a long line may break, useful for languages written without spaces.
- Zero-width joiner (U+200D) glues characters together. It builds emoji sequences, for example a family emoji is several people emoji joined by U+200D, and it shapes letters in some Indic scripts.
- Zero-width non-joiner (U+200C) stops letters from joining, which is needed to spell some Persian, Hindi and other words correctly.
- Word joiner (U+2060) prevents a line break at that point.
- Soft hyphen (U+00AD) appears as a hyphen only if a word has to break at the end of a line.
- Byte order mark (U+FEFF) marks the byte order of a file. It sometimes ends up inside text when a file saved with a BOM is joined to another.
- Direction controls such as the left-to-right mark (U+200E), right-to-left mark (U+200F) and the embedding, override and isolate controls mix left-to-right and right-to-left text.
- Non-breaking and special spaces such as U+00A0, the en and em spaces and the ideographic space (U+3000) control spacing and wrapping.
Why hidden characters cause trouble
Text copied from web pages, PDFs, word processors and chat apps often carries these characters without you knowing. The usual symptoms:
- Two strings look identical but a search, a match or a spreadsheet lookup says they differ.
- A password, username or code is rejected even though you typed it correctly.
- A program reports an "unexpected character" at a place that looks empty.
- A word is shown with a gap, or will not split at the end of a line.
- A file starts with a stray character that breaks a script or a CSV header.
- Text appears in the wrong order because of a direction control.
Direction controls deserve extra care. The right-to-left override (U+202E) can make a file name or a link look different from what it really is, which is why security guidance flags them. Cleaning them out of pasted text is sensible.
How to use the detector and cleaner
- Paste your text into the box. The count shows how many hidden or unusual characters were found, and a table lists each code point with its name and how many times it occurs.
- Look at the preview. Every hidden character is shown as a small tag with its code point, for example U+200B, coloured by type, so you can see exactly where it sits.
- Choose what to remove in the clean-up options. The defaults remove zero-width characters, the byte order mark, soft hyphens, direction controls, blank fillers, control characters and tag characters, and turn special spaces into normal spaces.
- Copy the cleaned text.
Two options are off by default on purpose. ZWJ and ZWNJ are left alone because removing them breaks emoji sequences and spelling in several scripts. Variation selectors are left alone because removing them can turn a coloured emoji into a plain symbol. Tick them only if you are sure.
The optional Unicode NFC step normalises equivalent character sequences, for example a letter followed by a combining accent, into the single composed character. It helps when two strings should match but are built differently.
What counts as hidden here
The tool flags characters from these groups: zero-width and invisible format characters, blank filler characters, soft hyphens, direction controls, space separators other than the ordinary space, line and paragraph separators, control characters other than tab, line feed and carriage return, variation selectors and Unicode tag characters. Ordinary letters, digits, punctuation, tabs and line breaks are never flagged.
Using it responsibly
This is a text cleaner. Use it to fix your own copied text, clean data and code, and understand what is inside a string. Removing invisible characters does not change what a text means or who wrote it, and it should not be relied on to hide how a text was produced. Many sites also prohibit blank or misleading names, so check the rules of the place where you paste.
Worked example
Paste the sample text from the button. The sentence contains a zero-width space inside "Hello world", a no-break space, a soft hyphen inside "this is", another zero-width space inside "test", a byte order mark, a right-to-left override with its closing pop, and an ideographic space. The detector lists each code point and count, and the cleaned version reads as a normal sentence with plain spaces. Press the Insert sample text button to see it.
Frequently asked questions
What is the Hangul filler character U+3164?
It is a Korean-script filler letter that most fonts draw as blank space. Because it counts as a letter rather than a space, it is often used to make a name look empty. Some sites strip it or refuse it.
What is the difference between U+3164 and U+2800?
U+3164 is the Hangul filler, a letter-like character. U+2800 is the Braille pattern blank, a symbol with no dots. Both look empty, and each is accepted in different places.
How do I find hidden characters in my text?
Paste the text into the detector. It lists every invisible or unusual character with its code point and name, and shows each one as a tag in the preview so you can see its position.
How do I remove zero-width spaces from text?
Paste the text, leave "Remove zero-width characters" ticked, and copy the cleaned text from the output box. The tool also removes the byte order mark, soft hyphens and direction controls by default.
Is it safe to remove ZWJ and ZWNJ?
Not always. The zero-width joiner builds emoji sequences and the non-joiner is needed in languages such as Persian. That is why both are kept unless you tick the option to remove them.
Why does my text have a no-break space?
Web pages and word processors insert U+00A0 so that two words stay on the same line. It looks like a normal space but is a different character, so it can break searches and imports. The cleaner turns it into a plain space.
Is my text uploaded?
No. Detection and cleaning run in your browser and nothing is sent to a server.
Will this work on a phone?
Yes. The copy buttons use your browser's clipboard. If a browser blocks it, select the text and copy it manually.
Sources & Methodology
- Unicode Standard Annex #9: Bidirectional Algorithm — direction marks, embeddings, overrides and isolates
- Unicode chart: General Punctuation (U+2000 to U+206F) — zero-width space, joiners, word joiner, invisible operators and special spaces
- Unicode chart: Braille Patterns (U+2800 to U+28FF) — Braille pattern blank, U+2800
- Unicode chart: Hangul Compatibility Jamo (U+3130 to U+318F) — Hangul filler, U+3164
- Unicode chart: Tags (U+E0000 to U+E007F) — tag characters
- Character names are the official Unicode names. The notes on where each character is accepted are general descriptions, not tested guarantees, and were not verified per website or app.
Data accurate as of: November 2026