Sensitive Data Redactor

About

Sensitive Data Redactor finds and removes sensitive personal information from web pages and PDF documents without ever sending your data anywhere. Every operation — scanning, detection, redaction, export — runs entirely within your browser, on your machine. Nothing is transmitted to any server.

Why this matters
Most redaction tools on the web ask you to upload your document. The moment you do that, you have lost control of it. You do not know who can access it on the other end, how long it is retained, whether it is used for training AI systems, or what the provider's breach history looks like. For legal documents, financial records, medical files, or anything commercially sensitive, that is an unacceptable trade-off. Uploading a confidential document to redact it is a contradiction in terms.

Sensitive Data Redactor takes the opposite approach. Your documents are processed locally by the same PDF engine your browser already uses. No account required. No file size limits imposed by a server. No usage logs. No analytics on what you redacted or when.

PDF redaction
Open any PDF directly in the built-in viewer. The extension scans the selectable text layer, highlights every detected item for your review, and — once you confirm — burns solid black boxes permanently into the document. The redacted PDF is then saved directly to your Downloads folder. The original text is not recoverable from the exported file. This works on any text-based PDF: court filings, contracts, financial statements, correspondence.

Web page redaction
On any web page, the scanner highlights detected items inline. You review, dismiss anything that is not sensitive, and apply. The page text is replaced with solid black blocks. From there, use the built-in Save as PDF function to export a clean, self-contained PDF of the redacted page — no uploading, no third-party conversion.

What it detects
Detection runs locally against thirteen categories, all of which can be individually enabled or disabled:

Social Security Numbers — US SSNs with format validation to minimise false positives
Email addresses — full RFC-compliant detection including subdomains and plus-addressing
Phone numbers — US and international formats including extensions
Credit card numbers — Visa, Mastercard, Amex, Discover via Luhn-prefix matching
EIN / Tax IDs — US employer identification numbers (XX-XXXXXXX format)
IP addresses — both plain and embedded in URLs
Dates — numeric and written formats across multiple conventions
URLs — full web addresses including paths and query parameters
Physical addresses — street-number-plus-suffix pattern covering common US/UK formats
Passport numbers — ICAO-standard formats covering US, UK, and most international variants
IBAN / Bank account numbers — all country variants of the international standard
Person names — detected using a locally-stored first-name and surname database that ships with the extension and never phones home
Custom keywords — add any word, phrase, or pattern you define; these are also stored locally

Grouped results and bulk dismissal
After scanning, results are grouped by type and by unique value. If a name appears forty times in a document, it shows as a single row with a count — one click dismisses or restores all instances of it together. You can also switch to an individual view to handle each occurrence separately. Nothing is permanently removed until you explicitly apply.

Sessions
Redaction decisions can be saved as sessions tied to a URL. Return to the same page later and restore the session — the extension re-scans and re-applies your previous choices automatically, respecting anything you had dismissed. Sessions are stored in your browser's local database.

Re-scan without limits
Adjust your category settings or add a custom keyword and re-scan as many times as you need. There are no usage limits, no scan credits, and no paywall. The name database can be refreshed on demand.

Privacy by design, not by policy
Privacy by policy means a company promises not to misuse your data. Privacy by design means the data never reaches them in the first place. Sensitive Data Redactor has no backend. There is no server that could be subpoenaed, breached, or sold. There is no account that could be de-anonymized. The extension cannot phone home about your documents because it has nowhere to call.