Quickstart /
Redact European PII in one function call. Choose your language and get running in 30 seconds.
Python
1. Install
$ pip install euredact2. Redact
from euredact import redact
result = redact("Mijn BSN is 111222333 en IBAN NL91ABNA0417164300.")
print(result.redacted_text)
# "Mijn BSN is [NATIONAL_ID] en IBAN [BANK_ACCOUNT]."
print(result.detections)Node.js
1. Install
$ npm install euredact2. Redact
import { redact } from "euredact";
const result = redact("Mijn BSN is 111222333 en IBAN NL91ABNA0417164300.", {
countries: ["NL"],
});
console.log(result.redactedText);
// "Mijn BSN is [NATIONAL_ID] en IBAN [BANK_ACCOUNT]."
console.log(result.detections);What Gets Detected
The SDKs detect 27 PII entity types across 31 European countries with 99.72% recall and 99.82% precision, backed by 346 pattern definitions and 44 checksum validators. Those figures come from a generated evaluation set of 152,300 records with the optional countries parameter supplied. Since 0.3.2 that parameter scores a detection rather than gating it: every pattern runs either way, and blind detection — no country declared — scores 99.50% recall and 99.59% precision. Date-of-birth detection is excluded and sits at 62.76% recall by design; bare dates are deferred to the AI layer.
Those 27 are the Rules Engine layer — everything with a format regular enough to match and, often, a checksum to confirm. The forthcoming AI layer adds 13 more for the categories that have no fixed shape: names, addresses, job titles, and the special-category data covered by GDPR Article 9. What euRedact detects lists all 40, what each one covers and deliberately does not, and which layer is responsible.