Skip to content

datenlabor-bmz/auto-redact

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

stability-wip

⚠️ DISCLAIMER: This software is currently in development and not yet ready for production use. Use at your own risk and always verify redactions manually.

⬛️ AutoRedact

A tool for redacting sensitive information from PDF documents using AI assistance.

Features

  • Upload and view PDF documents
  • Manual redaction by highlighting text
  • AI-powered detection of sensitive information
  • Export redacted PDFs

For a local-first version for the privacy-paranoid, see the SecuRedact fork.

Setup

  1. Create a .env file with your Azure OpenAI credentials:
AZURE_OPENAI_API_KEY=your_key
AZURE_OPENAI_API_BASE=your_base_url
AZURE_OPENAI_API_VERSION=your_version
  1. Build and run with Docker:
docker build -t auto-redact .
docker run -p 8000:8000 --env-file .env --rm auto-redact

Or run locally:

# Frontend
cd frontend
npm install
npm run dev

# Backend
cd backend
uv run python -m uvicorn main:app --reload

Note: When using npm run dev, loaded / generated highlights will not be displayed instantly, but this does work with the built version, which can also be started using npm run preview.

License

MIT License (c) BMZ Data Lab / David Pomerenke

The frontend is based on an example from react-pdf-highlighter, MIT License.

About

Redact sensitive information from PDFs with AI support

Resources

Stars

Watchers

Forks