Client-Side Sensitive Text Sanitizer

Paste or type text, then sanitize likely confidential or sensitive information in one click. This tool uses layered pattern matching, secret heuristics, and custom term matching to redact or mask sensitive values while preserving the surrounding text as much as possible.

Runs entirely in your browser No server calls No persistence after refresh Heuristic detection — manual review required
⚠️ Sanitization is heuristic. It can miss items or over-redact text, especially for names, companies, addresses, and ambiguous identifiers. Review the sanitized output and detection log before sharing.

Raw text input

Paste text below, choose a sanitization mode, add any custom confidential terms, then click Sanitize.

Everything is processed locally in this page. Refreshing the page clears the current session.
Use this for project names, internal codewords, brands, client names, device names, environment labels, or any other extra terms you want sanitized everywhere they appear.
Ready. Enter text and click Sanitize.

Sanitized output

This output stays in memory only for the current page session. Use the review panel to confirm what was changed.

Sanitized text will appear here after you run the sanitizer.
Tip: full redaction is safer for sharing externally. Partial masking is useful when reviewers still need some contextual visibility without exposing the full value.

Detection review

Each detection shows the category, a safe preview of the matched value, and the replacement used in the output.

Detections 0
Categories 0
Mode —
No detections yet. Run the sanitizer to see what was identified and how it was handled.
The detector set is deliberately modular so new rules can be added later. Structured identifiers tend to be more reliable than names, companies, and free-form addresses.