NoteLink is a premium, high-performance platform for note-taking, analysis, and research. It transforms the way you capture knowledge from YouTube, the Web, and personal notes by leveraging SQLite FTS5, Model Context Protocol (MCP), and Local/Cloud AI.
- Instant Search (FTS5): Near-instant full-text search across thousands of notes and massive video transcripts using SQLite's virtual table indexing.
- Delta Sync Protocol: Optimized backend synchronization that only updates modified records, reducing disk I/O and saving battery life.
- Lazy Loading Architecture: Metadata-first note retrieval that keeps the UI responsive even with large datasets.
- Summarization Memory: AI-generated summaries are cached in the database. Open a note, and your summary appears in milliseconds without re-invoking the AI.
- Permanent URL Analysis: Summaries generated from YouTube or Web links are now permanently stored in the DB (migrated from volatile LocalStorage).
- Smart Invalidation: The system intelligently detects content changes to determine when a summary needs to be refreshed.
Export your AI-extracted tasks and insights directly to your favorite tools:
- Notion: Database-synced research archives.
- Trello: Automated task boards for planning.
- Google Calendar: Timeline-managed deadlines.
- Obsidian: Local-first markdown knowledge base.
- AI Summarizer & Narrator:
- 📄 Text Summary: Deep, structured professional overviews.
- 🎧 Voice Narration: Auditory-optimized scripts with real-time text-to-speech.
- Automated Metadata Extraction: Uses
yt-dlpandBeautifulSoupto fetch actual video/article titles and metadata automatically. - Multilingual Support: Smart language handling that preserves Arabic for text summaries while providing High-Quality English narration.
- Frontend: React + Vite, Framer Motion, Lucide Icons.
- Backend: FastAPI, Python (Uvicorn).
- Database: SQLite with FTS5 (Full-Text Search) and Delta Sync.
- AI Engine:
- Local: Ollama (Gemma 4/llama3).
- Cloud: Google Gemini Pro (via API).
- Transcription: OpenAI Whisper (Local).
- Protocol: Model Context Protocol (MCP) for standardized tool execution.
- Node.js: For the React frontend.
- Python 3.10+: For the FastAPI backend.
- Ollama: (Optional) For local-only AI processing.
- FFmpeg: Required for transcribing media files.
cd backend
pip install -r requirements.txt
python main.pynpm install
npm run dev- Device-Local Data: Your database (
notes.db) and secrets (.env) are automatically ignored by Git to ensure your research stays private on your machine. - Unified Settings: Configure AI providers, API keys, and cloud integrations via the native settings dashboard.
Developed for researchers who value privacy, performance, and speed.