Hakkutsu (発掘, "excavation/discovery") is a Chrome browser extension that transforms any webpage into a Japanese learning environment. It detects Japanese text, provides instant analysis with tokenization, dictionary definitions, furigana, and JLPT levels, and lets you export vocabulary to Anki with one click.

• 🔤 Japanese Text Detection — Automatically identifies Japanese text on any webpage
• 📖 Deep Analysis — Tokenization, readings, POS tags, dictionary definitions, example sentences
• 🏷️ JLPT Levels — Instant N5–N1 classification for every word
• 📺 YouTube Subtitles — Extract and analyze Japanese subtitles from videos
• 🖼️ OCR Support — Extract Japanese text from images and manga panels
• 📇 Anki Export — One-click flashcard creation via AnkiConnect
• 🔐 Cloud Sync — Save vocabulary and settings with Firebase
• 🤖 AI Difficulty Scoring — Transformer-based sentence difficulty classification (Phase 2)

Hakkutsu/
├── extension/     # Chrome Extension (Plasmo + React + TypeScript)
├── backend/       # API Server (FastAPI + Python)
├── ml/            # ML Pipeline (PyTorch + Transformers) — Phase 2
└── shared/        # Shared type definitions

• Node.js 18+
• pnpm 8+
• Python 3.11+
• Chrome browser

cd extension
pnpm install
pnpm dev

Then load the extension in Chrome:

1. Navigate to chrome://extensions
2. Enable "Developer mode"
3. Click "Load unpacked"
4. Select extension/build/chrome-mv3-dev

cd backend
python -m venv .venv
.venv/Scripts/activate  # Windows
pip install -r requirements.txt
python -m app.main

The API will be available at http://localhost:8000/docs.

This project is licensed under the MIT License — see the LICENSE file for details.

• JMdict by the Electronic Dictionary Research and Development Group
• SudachiPy by Works Applications
• Plasmo browser extension framework
• AnkiConnect by FooSoft Productions