PII Redactor

How to use

1. 1Paste any text containing personally identifiable information into the Original text area on the left, or click Load sample to try it out.
2. 2On the right, toggle which built-in rules to apply (emails, phones, credit cards with Luhn check, SSNs, IBANs, IPs, MACs, UUIDs, JWTs, AWS keys, ZIPs, URLs).
3. 3Pick a placeholder style: Named ([EMAIL]), Numbered ([EMAIL_1]), a fixed string of your choice, or stars that preserve the original length.
4. 4Add a custom regex pattern for anything organization-specific (employee IDs, ticket numbers, internal hostnames). Errors are reported inline.
5. 5Verify the result in the Matches table (toggle Show originals if you need to review them) and click Copy redacted to grab the cleaned text.

About this tool

PII Redactor scrubs personally identifiable information out of any text, locally in your browser. Paste a log file, a support ticket, a chat transcript, a CSV column, or a prompt you want to send to an LLM, then pick which categories to mask. Built-in rules cover email addresses, phone numbers (7 to 15 digit international formats), credit card numbers (13 to 19 digits, validated with the Luhn checksum to avoid false positives on plain digit runs), US Social Security Numbers (with invalid-area filtering), IBANs (validated with the mod-97 check), IPv4 and IPv6 addresses, MAC addresses, UUIDs, JWT tokens, AWS access key IDs, US ZIP codes, and URLs. Choose how matches are replaced: a named tag like [EMAIL], a numbered tag like [EMAIL_1] so distinct values stay distinguishable in the cleaned output, a fixed string of your choice, or stars that preserve the original length and the first and last visible characters. Add your own regex patterns for organization-specific identifiers (employee IDs, ticket numbers, internal hostnames, API key formats) and they slot into the same pipeline with a toggleable enable switch, custom name, custom placeholder tag, and inline regex error messages. The match table shows every redacted value with its position so you can verify nothing important was caught by accident, and a Show originals toggle keeps the table useful for review without printing originals by default. Useful for preparing logs for bug reports, sanitizing transcripts before sharing with vendors, redacting screenshots and OCR output before posting them publicly, anonymizing exported CSVs, and scrubbing prompts before sending them to a third-party AI. Every match runs locally with regex matching plus checksum validation where applicable. Nothing is uploaded, logged, or sent anywhere; this is exactly the use case where it matters that no signup, no upload, and no telemetry is the literal truth. Treat the result as a best-effort scrub and not a substitute for a formal data loss prevention review.

Free to use. Works in your browser. No signup, no login.