Presidio

Presidio is an open-source SDK from Microsoft for detecting and anonymizing personally identifiable information (PII) in text, images, and structured data. It identifies entities like names, locations, credit cards, SSNs, and phone numbers using NLP, regex, rule-based logic, and checksums, then redacts or anonymizes them.

It installs via pip, Docker, or Kubernetes, and supports custom recognizers and de-identification of DICOM medical images. Often used as the PII layer inside a larger guardrails stack. MIT licensed.