A clone of Cluely, because it didn't work on Windows when I tried it.
So I built it from scratch faster than they could fix it.
Updated Demo
Updated Demo 2
Old Demo 1
Old Demo 2
- 🕵️♂️ Invisible to screen capture tools
- 👀 Can see your screen
- 🎤 Can hear you alongside your device, including others in meetings (this is a good thing)
- 🔄 Two modes: chat, voice. Both have access to your screen
- Almost instantaneous response in voice mode
- ⌨️ Shortcut enabled: quickly hide with Ctrl+Space, Send messages with Ctrl+Enter
- 🛠️ Improve model behaviour using system prompts
- 📦 Ship a standalone .exe
- 🤖 Currently only interfaces with Gemini
- 🪟 Aimed at Windows; unsure how it will perform on macOS, Linux
| Shortcut | Action |
|---|---|
| Ctrl+Space | Toggle invisibility (show/hide main and chat windows) |
| Ctrl+Enter | Open chat window / submit input in chat mode |
| Esc | Cancel chat or close windows |
| Mode | Input | Description |
|---|---|---|
| Chat | Keyboard | Type prompts; captures a screenshot and returns LLM response. |
| Voice | Microphone | Streams live audio & screenshots; responses appear in real-time. |
- Node.js (v18 or higher)
- npm, yarn, pnpm, or bun
Clone the repository and install dependencies:
# Clone the repository
# Install dependencies
npm installSet your Gemini API key in the .env file:
VITE_GEMINI_API_KEY=*****Start the development server:
npm run devThis will start Clonely with hot-reload enabled so you can see changes in real time.
Build the application for your platform:
# For Windows
npm run build:win
# For macOS
npm run build:mac
# For Linux
npm run build:linux
# Unpacked for all platforms
npm run build:unpackDistribution files will be located in the dist directory.