🛡️ Redac

A lightweight PII (Personally Identifiable Information) moderation MVP designed to sanitize sensitive data before it reaches LLM APIs.

🚀 Key Features

• Multi-Language Support: High-accuracy detection for English and French using spaCy Large models.
• Double-Pass Protection: Combines NLP-based detection with expert Regex patterns for PII coverage.
• Expert French Recognizers: Built-in support for French-specific data: SIRET, NIR (Social Security), IBAN, and addresses.
• Balanced Anonymization: Preserves job titles and document structure to keep texts readable.
• Minimal Dashboard: React-based UI with Risk Assessment visualization.
• Custom Theme UI: Switch between Premium, Minimal Light, and Deep Midnight modes.

---

🎥 Demo

Check out Redac in action:

Click to watch the full demo on YouTube

---

🛠️ Architecture

1. Core API (/api): FastAPI server powered by Microsoft Presidio.
2. Web Dashboard (/ui): React + Vite + Tailwind CSS.

---

📦 Getting Started (Docker)

Quick Start with Makefile

If you have make installed, you can use these shortcuts:

make build   # Build containers
make up      # Start Redac
make logs    # View logs
make test    # Run tests

Manual Docker commands

docker compose up --build

• API: http://localhost:8000
• UI Dashboard: http://localhost:5173

---

🔌 API Usage

You can interact directly with the Redac API using curl.

Redact Text

Request:

curl -X POST http://localhost:8000/redact \
     -H "Content-Type: application/json" \
     -d '{"text": "My name is John Doe, call me at 06 12 34 56 78."}'

Response (JSON):

{
  "original_text": "My name is John Doe, call me at 06 12 34 56 78.",
  "redacted_text": "My name is <PERSON>, call me at <PHONE_NUMBER>.",
  "detected_language": "en",
  "entities": [
    {
      "type": "PERSON",
      "text": "John Doe",
      "score": 85,
      "start": 11,
      "end": 19
    },
    {
      "type": "PHONE_NUMBER",
      "text": "06 12 34 56 78",
      "score": 95,
      "start": 32,
      "end": 46
    }
  ]
}

---

🧪 Quality Assurance

Using Makefile

make test

Manual Docker

# Run the test suite
docker compose run api python tests/verify_all.py

---

📄 License

MIT - Created for secure LLM workflows.