String obfuscator
Turn text into something hard to read at a glance — zero-width characters, homoglyphs, reversed Base64, character shifting or RTL marks — and get the matching way to restore it.
Runs in your browserEvery computation happens in your browser — your data never leaves this device.
Obfuscated
Hello, UniKit!
H\u200be\u200bl\u200bl\u200bo\u200b,\u200b \u200bU\u200bn\u200bi\u200bK\u200bi\u200bt\u200b!\u200b
1428561400Restore
What this tool does
- Demonstrate text that looks identical but is not: after inserting zero-width characters the glyphs are unchanged while the code point and byte counts grow, so search and comparison break.
- Build security training material — swap UniKit for a Cyrillic homoglyph version and show how lookalike domains and fake accounts are disguised.
- Lightly hide a plain string (reversed Base64, character shift) so it is not readable at a glance in a screenshot or chat, with the matching restore function on the same page.
- Debug why pasted text from a web page will not match: check how many zero-width and directional marks are hiding inside it.
Example
Input
UniKit (method: Base64 + reverse)
Output
0l2Sp5WV
UniKit is UTF-8 encoded and Base64-encoded to VVcpaUtpblU=, then reversed. Reverse 0l2Sp5WV back to VVcpaUtpblU= and Base64-decode to recover UniKit. With the character shift method, a shift of 3 turns UniKit into XqlNlw.
Frequently asked questions
Can text with zero-width characters be restored?
Yes. The inserted characters are the fixed code points U+200B, U+200C, U+200D and U+FEFF, so restoring simply strips them and leaves the visible text intact. Directional marks (U+202A–U+202E, U+2066–U+2069) are stripped as well.
Does character shift work on Chinese text?
No, Chinese characters are left untouched. Shifting only applies inside the printable ASCII range 33–126 and wraps within it, so 你好A shifted by 3 becomes 你好D. The shift amount ranges from 1 to 93.
Can homoglyph restore damage genuine Cyrillic text?
Yes. Restoring maps every Cyrillic or Greek lookalike back to ASCII unconditionally, so real Russian text gets rewritten into Latin letters too. It is meant for recovering deliberately disguised strings, not as language detection.
Why does Base64 + reverse restore say "not a valid result of the current method"?
Restoring reverses the string and then Base64-decodes it as standard Base64 with a strict UTF-8 check. If the text was truncated in transit, line breaks were stripped, or a URL-safe alphabet (- and _) was used, decoding fails. Handling it in the dedicated Base64 tool is more forgiving in that case.
Can I use these methods as encryption?
No. All five modes are reversible encodings or substitutions whose logic is printed on this very page, so anyone can undo them. Use real encryption such as AES when secrecy matters; obfuscation is for demos, redaction or a temporary visual barrier only.
Keywords：混淆obfuscate隐藏文本zero width零宽字符同形字homoglyphrtlbase64