IAC-099 Module 2: Privacy Filter

Regex-based PII detection and filtering for sensitive content.

Purpose

Scans transcribed text for personally identifiable information (PII) using regex patterns. User manually selects what to keep or remove. No AI needed. Works offline.

Input/Output Contract

Input: transcript.md - Raw transcript from Module 1 Output: filtered.md - Privacy-filtered markdown

Tech Stack

• Framework: Tauri 2
• Frontend: Svelte 5
• PII Detection: Regex patterns (pure pattern matching)
• Database: SQLite (local storage for patterns)
• Platform: macOS (Apple Silicon M2)

Prerequisites

• macOS with Apple Silicon (M2/M3)
• Node.js 20 (via nvm)
• Rust (latest stable)

Installation

# Use Node 20
nvm use 20

# Install dependencies
npm install

Development

# Start dev server
npm run tauri:dev

# Run tests
npm test

# Lint code
npm run lint

Building

# Build production app
npm run tauri:build

Usage

1. Load transcript.md from Module 1
2. App highlights PII using color codes:
   • Red: Private (SSN, credit cards, passwords)
   • Orange: Personal (names, emails, phone numbers)
   • Yellow: Business (company names, addresses)
   • Green: General (safe content)
3. User deselects (unchecks) sensitive content
4. Click "Export Filtered" to save filtered.md

PII Detection Categories

Private (Highest Sensitivity)

• Social Security Numbers
• Credit card numbers
• Bank account numbers
• Passwords/credentials
• Medical record numbers

Personal

• Full names
• Email addresses
• Phone numbers
• Physical addresses
• Dates of birth

Business

• Company names
• Business addresses
• Business phone numbers
• Tax IDs

General

• Everything else (safe to share)

Configuration

• Custom regex patterns
• Sensitivity levels
• Auto-filter options
• Pattern library management

Source Code References

Based on working implementation from:

• /Users/kjd/01-projects/IAC-031-clear-voice-app/test-privacy-scanner.js

Success Criteria

• ✅ Loads markdown
• ✅ Highlights PII with regex
• ✅ User can filter content
• ✅ Outputs clean markdown
• ✅ Works offline
• ✅ No AI dependencies

Architecture

src/
├── components/      # Svelte UI components
│   ├── Highlighter.svelte
│   └── FilterPanel.svelte
├── lib/
│   ├── patterns/   # Regex pattern definitions
│   └── scanner.js  # PII detection logic
└── tauri/          # Rust backend
    └── storage/    # Pattern storage

Privacy & Security

• All processing happens locally
• No data leaves the device
• Pure regex matching (no AI/ML)
• No cloud API calls
• User has full control over filtering

License

[To be determined]

Contributing

See main project repository for contribution guidelines.