Reveal invisible and unusual Unicode code points with positions, names and a readable annotated preview.
Files are read and cleaned locally. Download a combined text result when processing finishes.
Paste text to see a transparent report.
Highlighted labels show unusual or transformed characters at their original positions.
Two strings can look identical while containing different code points. The inspector identifies control characters, format markers, Unicode spaces, smart punctuation and non-ASCII symbols.
The default inspector does not need to rewrite the text. Use the report to understand a failed comparison or validation issue, then enable only the cleanup rule you need.
A BC
A B C
The report identifies an NBSP and a zero-width space.
It summarises ordinary ASCII and lists unusual or invisible code points that are most useful for debugging.
The browser interface provides the cleaned text and a visible report; report export can be added without sending text to the server.
No. Most languages require non-ASCII characters. The report is descriptive, not a warning that all Unicode should be removed.
Not every unusual character is an AI watermark, and deleting Unicode cannot guarantee removal of a model-level statistical watermark.
Read guide