Settings

Configure which types of PII are detected and how the app behaves.

Detection Engine

Controls how PII is identified in your text

NER Model (names, orgs, locations)

Uses a local BERT model (~30 MB, cached after first load) to detect unstructured PII like person names, organizations, and locations.

Auto-mask on paste

Automatically run detection when you paste text into the Mask editor, without requiring a click.

PII Categories

Toggle which types of personal data are detected and masked. 13 / 13 active

Identity

Name<PERSON_1>

Full names, first names, last names

Org<ORG_1>

Company, institution, and team names

Location<LOC_1>

Cities, countries, landmarks

Address<ADDRESS_1>

Street addresses with house numbers

Contact

Email<EMAIL_1>

Email addresses in any format

Phone<PHONE_1>

Phone numbers, including international formats

Government IDs

SSN<SSN_1>

US Social Security Numbers (XXX-XX-XXXX)

Date of Birth<DOB_1>

Dates that appear to be dates of birth

Financial

Card<CC_1>

Credit/debit card numbers with Luhn validation

IBAN<IBAN_1>

International bank account numbers

Technical

API Key<API_KEY_1>

API keys, tokens, and secrets (sk-, ghp_, AKIA…)

Wallet<CRYPTO_1>

Bitcoin, Ethereum, and other wallet addresses

IP Address<IP_1>

IPv4 and IPv6 addresses

These settings are stored only in your browser's local storage — they contain no PII and are never transmitted anywhere.Detection and masking always happen 100% in your browser.