Presidio
Open-source Python framework to find and anonymize personal data in text, images and structured data.
- Stars
- 11.6k
- Forks
- 1.4k
- Licence
- MIT
- Language
- Python
- Last update
- 5 days ago
- Type
- Library
What it is
Presidio is a Python framework for detecting sensitive personal information (PII) and then redacting, masking, or anonymizing it. According to its description, it works across text, images, and structured data, and combines NLP, pattern matching, and customizable pipelines. People use it to remove personal details from data before storing, sharing, or processing it.
- Detect names, phone numbers, or other PII in free text before logging it
- Redact or mask sensitive fields in structured records
- Build a custom pipeline that combines NLP models with pattern rules
Great for
- Developers adding PII detection to Python data pipelines
- Teams that need to de-identify text before sending it to other services
- Data engineers preparing datasets that must not expose personal details
- Privacy-focused teams wanting a self-hosted, inspectable tool under the MIT license
Not ideal for
- People who need a no-code, point-and-click tool, since the project is a Python framework and the README provided here does not describe a graphical app
- Anyone expecting guaranteed detection; the README excerpt provided does not document accuracy figures, so results should be tested on your own data
How to get it
- 1
Install with a package manager
You'll need: Python
Clone the repository
git clone https://github.com/data-privacy-stack/presidio - 2
Run it
- Clone the repository and read the README for setup instructions
- Install the Python dependencies as described in the project docs
- Run a small test on sample text to see what gets detected
- Adjust the pipeline to fit your data and re-test
- 3
Get the most out of it
- Test detection on representative samples of your own data before relying on it
- Customize pipelines to combine NLP and pattern-based recognizers
- Check the project homepage for documentation: https://presidio.dataprivacystack.org
Downloads
Strengths
- MIT license permits broad use
- Covers several kinds of data (text, images, structured) according to its description
- Customizable pipelines let you adapt detection to your own needs
Trade-offs
- The README excerpt provided was truncated, so specific features and setup details could not be verified here
- Detection of sensitive data is never perfect; you must validate results for your use case
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| presidio (this) | 11.6k | Library | intermediate | 5 days ago | Developers adding PII detection to Python data pipelines |
| LEANN | 13.8k | Library | intermediate | 4 days ago | Developers building a private RAG prototype on a personal machine |
| quivr | 39.6k | Framework | advanced | 4 days ago | Developers building a search or monitoring pipeline over a stream of documents |
| web-check | 35.3k | App | intermediate | 4 days ago | Developers checking how a site is configured |
| hosts | 31.4k | Other | beginner | 4 days ago | People who want to block ads and known malware domains across a whole device or home network |
| vit-pytorch | 25.6k | Library | intermediate | 5 days ago | Developers who want to build an image classifier in PyTorch |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.