Unicode Security Scanner
Inspect text locally for deceptive look-alikes, mixed writing systems, bidirectional controls, invisible characters, controls, noncharacters, and normalization surprises. No text leaves your browser.
Findings
Run a scan to see findings.
Per-code-point inspection
| Position (0-based) | Character | Code point | Script / category | Security notes |
|---|
Explicit cleanup preview
These actions only generate a preview. They never alter the input. Review the result before copying it.
How to interpret this scan
Heuristic limits: this is a focused client-side check, not a full implementation of Unicode Technical Standard #39 and not proof that text is malicious or safe. Mixed scripts are normal in multilingual text; confusable coverage is intentionally limited to common Latin/Greek/Cyrillic look-alikes; script classification uses broad browser Unicode properties; and normalization can change legitimate text. For security-sensitive identifiers, also enforce an application-specific allowlist and compare against a maintained Unicode confusables dataset on the server.