How to use Unicode Code Point Inspector
- Choose the input kind, then paste literal text, hexadecimal code points or supported Unicode escapes.
- Choose Code points and encodings or Grapheme groups in Result table.
- Review the names, encoded bytes and any surrogate or script hints. Download JSON for the complete records.
Example: Unicode Code Point Inspector
See why A, an emoji and an accented e contain four code points.
Options
- Text mode
- The tool inspects the entered sequence without normalization. A visible accented letter can consist of one code point or several.
- Code-point input
- Use the supported code-point form for explicit inspection. Names, blocks and scripts come from the pinned Unicode 17 dataset.
Supported inputs and limits
- Up to 4,096 UTF-16 input units or 1,024 entered code points.
- Names/categories/blocks/scripts use Unicode 17.0.0.
- Unpaired surrogates are flagged.
- Literal text is inspected without normalization.
- Escape mode explicitly decodes supported sequences before inspection; the input and decoded result should be reviewed separately.
More details
Optional browser grapheme grouping describes visible clusters.
Mixed-script hints are review aids, not confusable detection, glyph coverage or security certification.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
A code-point record does not describe a glyph
Fonts can combine or omit characters, and several code points can form one visible grapheme. Unpaired surrogates are flagged rather than interpreted as ordinary characters. Mixed-script hints call for review; they do not establish impersonation or safe rendering.
Questions about Unicode Code Point Inspector
Is a character always one code point?
No. A displayed symbol can contain several code points.
Does a mixed-script hint prove spoofing?
No. Many legitimate strings use multiple scripts; review the actual text.
Does this normalize the text?
Literal mode does not normalize entered text. Escape mode decodes the selected escape syntax before inspection; it does not normalize the decoded text.