Jot

An open-source, privacy-first AI meeting notepad for macOS

Keep your meeting transcripts local with AI-powered summaries. No account required.

Why Jot? Unlike cloud-based tools, Jot runs Whisper transcription locally on your Mac. No account required, no subscriptions, no data leaving your machine (except optional AI summaries).

Features

Apple Notes-Inspired UI - Clean, minimal interface with sidebar navigation and rich text editing
Live Transcription - Real-time speech-to-text during meetings using Whisper AI (runs locally, no cloud required)
Chat with Note - Ask questions about your notes or request AI to add/modify content
Privacy-First - All transcription happens on your Mac, no account/signup needed, SQLite database you control
Dual Audio Recording - Capture both system audio and microphone simultaneously via macOS ScreenCaptureKit
AI Summaries - Generate intelligent meeting summaries with 8 contextual prompts (action items, key decisions, Q&A, etc.)
Multi-Color Highlights - Organize your notes with 6 highlight colors (yellow, orange, pink, purple, green, blue)
Apple Notes Import - Drag notes from Apple Notes to Finder, then import .txt/.md/.html/.rtf files
Audio Playback - Optional embedded player to review meeting audio
Keyboard Shortcuts - Cmd+N (new note), Cmd+Backspace (delete), Cmd+R (toggle recording)
Open Source - MIT licensed, audit the code, contribute features, self-host if you want

Tech Stack

Frontend: React 19 + TypeScript + Vite 7 + Tailwind v4
Editor: Tiptap v3 (rich text with tables, lists, highlights)
Backend: Tauri v2 (Rust 1.93) with macOS ScreenCaptureKit
Database: SQLite via tauri-plugin-sql
Transcription: Whisper (whisper-rs, runs locally)
AI: Multi-provider support (OpenAI, Anthropic, DeepSeek, Moonshot)
State: Zustand

Installation

Requirements

macOS 10.15+ (Catalina or later)
Node.js 22.x (for development)
CMake (for building Whisper)
```
brew install cmake
```

Rust 1.93+ (for Tauri)

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

pnpm (package manager)
```
npm install -g pnpm
```

Development Setup

# Clone the repository
git clone https://github.com/yourusername/jot.git
cd jot

# Install Node 22 (using nvm)
nvm install 22
nvm use 22

# Install dependencies
pnpm install

# Start dev server
./dev.sh
# Or manually:
export MACOSX_DEPLOYMENT_TARGET=10.15
pnpm tauri dev

Production Build

export MACOSX_DEPLOYMENT_TARGET=10.15
pnpm tauri build

Note: Production builds currently have issues with whisper-rs CMake linking. Use GitHub Actions for releases or run in dev mode.

Usage

First-Time Setup

Grant Screen Recording Permission
- System Settings > Privacy & Security > Screen Recording
- Enable "Jot"
- Restart the app
Download Whisper Model
- Click Settings (⚙️) in sidebar
- Select a model (Tiny = 74 MB, Base = 142 MB, Small = 466 MB)
- Click "Download Model"
- Wait for download to complete
Add AI API Key (for summaries and chat)
- Choose your preferred provider in Settings:
  - OpenAI - Get key from https://platform.openai.com/api-keys
  - Anthropic - Get key from https://console.anthropic.com/settings/keys
  - DeepSeek - Get key from https://platform.deepseek.com/api_keys
  - Moonshot - Get key from https://platform.moonshot.cn/console/api-keys
- Paste key into Settings
- Your key is stored locally and only sent to your chosen provider

Recording a Meeting

Click Record button (or press Cmd+R)
Choose audio sources:
- System Audio - Capture speakers (Zoom, Meet, etc.)
- Microphone - Capture your voice
Speak normally - live transcription appears every 15 seconds
Click Stop to end recording
(Optional) Click Summarize for AI-generated summary

Importing from Apple Notes

In Apple Notes, select a note
File > Export as Plain Text... (or drag to Finder)
In Jot, click Upload (📤) icon in sidebar
Select .txt/.md/.html/.rtf files (batch supported)
Notes appear in "Imported" group

Keyboard Shortcuts

Shortcut	Action
`Cmd+N`	Create new note
`Cmd+Backspace`	Delete current note
`Cmd+R`	Toggle recording (start/stop)
`Cmd+B`	Bold
`Cmd+I`	Italic
`Cmd+U`	Underline

File Locations

Database: ~/Library/Application Support/com.jot.app/jot.db
Whisper Models: ~/Library/Application Support/com.jot.app/models/
Audio Recordings: ~/Library/Application Support/com.jot.app/recordings/ (if "Keep Audio" enabled)

AI Providers

Jot supports multiple AI providers for summaries and chat. Choose your preferred provider in Settings:

OpenAI - GPT-5.2, GPT-5-mini (default), GPT-5-nano
Anthropic - Claude 4.6 Opus, Claude 4.5 Sonnet (default), Claude 4.5 Haiku
DeepSeek - DeepSeek Chat (default), DeepSeek Reasoner
Moonshot - Kimi k2.5

Your API key is stored locally in SQLite and only sent to your chosen provider. No data is shared with other providers.

Getting API Keys

OpenAI: https://platform.openai.com/api-keys
Anthropic: https://console.anthropic.com/settings/keys
DeepSeek: https://platform.deepseek.com/api_keys
Moonshot: https://platform.moonshot.cn/console/api-keys

Switching Providers

Open Settings (⚙️ in sidebar)
Click your preferred provider logo (OpenAI / Anthropic / DeepSeek / Moonshot)
Enter your API key for that provider
Use "Summarize" or "Chat with Note" - requests go to your selected provider

Troubleshooting

"Model not downloaded" error

Go to Settings > Download Model
Wait for download to complete (progress shown in button)
Model file should appear in ~/Library/Application Support/com.jot.app/models/

Recording fails with permission error

Grant Screen Recording permission in System Settings
Restart Jot after granting permission
Click "Open Settings" button in error message for quick access

Live transcription not working

Check that Whisper model is downloaded
Ensure you're speaking loud enough (test with system voice memo)
Try switching to a larger model (Base or Small) for better accuracy

Build fails with CMake errors

Ensure CMake is installed: brew install cmake
Set deployment target: export MACOSX_DEPLOYMENT_TARGET=10.15
Clean build: cd src-tauri && cargo clean

Development

Project Structure

jot/
├── src/                        # React frontend
│   ├── components/             # UI components (Sidebar, Editor, Settings, etc.)
│   ├── store/                  # Zustand stores (notes, recording, settings)
│   ├── lib/
│   │   └── ai/                 # Multi-provider AI integration
│   │       ├── types.ts        # Shared AI types
│   │       ├── providers.ts    # Provider registry (OpenAI, Anthropic, DeepSeek)
│   │       ├── client.ts       # Universal API client
│   │       ├── summarize.ts    # Meeting summarization
│   │       └── chat.ts         # Chat with note
│   └── index.css               # Design system (CSS variables)
├── src-tauri/                  # Rust backend
│   ├── src/
│   │   ├── audio.rs            # ScreenCaptureKit audio recording
│   │   ├── transcribe.rs       # Whisper model download & transcription
│   │   ├── live_transcribe.rs  # Live chunked transcription
│   │   └── lib.rs              # Tauri commands
│   └── tauri.conf.json         # Tauri configuration
└── package.json                # Frontend dependencies

Design System

Primary Color: Orange (#FD3200) at 4-12% opacity for backgrounds, solid for CTAs
Typography: Native macOS fonts (-apple-system, SF Pro Text)
Spacing: Generous whitespace, 6-10px border radius
Aesthetic: Apple Notes-inspired (clean, minimal, professional)

Code Style

TypeScript: Strict mode enabled
Components: Functional components with hooks
State: Zustand stores (separate concerns)
Memoization: useMemo/memo for performance
Lazy Loading: Settings, RecordingOptions, ImportModal

Bundle Size

Frontend: 676 kB (214 kB gzipped)
Lazy-loaded: Settings (2.43 kB), RecordingOptions (1.06 kB)
Zero TypeScript/ESLint warnings

Testing

# TypeScript check
pnpm tsc --noEmit

# Lint
pnpm lint

# Build frontend
pnpm build

# Rust check
cd src-tauri && cargo check

Roadmap

Contributing

Contributions welcome! Please:

Fork the repository
Create a feature branch (git checkout -b feat/my-feature)
Commit changes with descriptive messages
Push to your fork and submit a pull request

License

MIT License - see LICENSE for details

Acknowledgments

Tauri - Cross-platform app framework
Whisper - Speech recognition
Tiptap - Rich text editor
ScreenCaptureKit - macOS audio capture

Support

Issues: https://github.com/yourusername/jot/issues
Discussions: https://github.com/yourusername/jot/discussions

Made with ❤️ for productive meetings

Name		Name	Last commit message	Last commit date
Latest commit History 59 Commits
.github		.github
src-tauri		src-tauri
src		src
.gitignore		.gitignore
.nvmrc		.nvmrc
APPLE_NOTES_IMPORT_PLAN.md		APPLE_NOTES_IMPORT_PLAN.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
PERMISSIONS.md		PERMISSIONS.md
README.md		README.md
SCREENSHOTS.md		SCREENSHOTS.md
dev.sh		dev.sh
eslint.config.js		eslint.config.js
index.html		index.html
jot-social.png		jot-social.png
package.json		package.json
pnpm-lock.yaml		pnpm-lock.yaml
request_mic_permission.sh		request_mic_permission.sh
resign.sh		resign.sh
tsconfig.app.json		tsconfig.app.json
tsconfig.json		tsconfig.json
tsconfig.node.json		tsconfig.node.json
vite.config.ts		vite.config.ts

License

h0ds/jot

Folders and files

Latest commit

History

Repository files navigation

Jot

Features

Tech Stack

Installation

Requirements

Development Setup

Production Build

Usage

First-Time Setup

Recording a Meeting

Importing from Apple Notes

Keyboard Shortcuts

File Locations

AI Providers

Getting API Keys

Switching Providers

Troubleshooting

"Model not downloaded" error

Recording fails with permission error

Live transcription not working

Build fails with CMake errors

Development

Project Structure

Design System

Code Style

Bundle Size

Testing

Roadmap

Contributing

License

Acknowledgments

Support

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases 1

Packages 0

Languages

Packages