Library / SDK
Guardrails & Structure
Presidio
1Detect 2and 3anonymize 4PII 5in 6text
the six Ws · specification
W1
Who
Maintained by Microsoft.
W2
What
A data-protection SDK that detects and anonymizes PII and sensitive entities in text and images.
W3
Where
A Python library and optional containerized services, self-hosted.
W4
When
Reach for it when scrubbing personal data before logging or sending to an LLM.
W5
Why
Customizable, extensible PII detection and de-identification out of the box.
W6
With
Uses spaCy and transformer NER models; integrates with custom recognizers.
W7
Watch
Maintained by Microsoft; open-source (MIT) and fully self-hostable. Runs locally with local models, so sensitive text need not leave your environment.
alternatives