Multidoro is a state-of-the-art, glassmorphic desktop Pomodoro application built with Electron, TypeScript, and HTML5/Vanilla CSS. It integrates stateless Gemini HTTP APIs to act as a real-time visual work coach—monitoring your active screen and speaking warnings out loud if you get distracted from your declared task.
| Main Dashboard | Analytics Logs & Settings | System Tray Menu |
|---|---|---|
 |
 |
 |
- 🧠 Real-Time AI Coaching: Uses stateless Gemini APIs (Gemini 3.5 Flash or Gemini 3.1 Pro) to analyze screenshots of your active screen against your declared task under a structured JSON schema.
- 🔊 Voice Feedback: Generates on-demand verbal scoldings (via Gemini 3.1 Flash TTS) using multiple selectable voices (
Zephyr,Puck,Aoede, etc.) to keep you strictly on-task. - 🎙️ Speech-to-Text Task Input: Dictate your current task hands-free using built-in speech recognition.
- ⏱️ Flexible Pomodoro Techniques:
- Traditional: 25m focus / 5m break.
- Ultradian: 50m focus / 10m break (ideal for deep work).
- Animedoro: 60m focus / 20m break.
- Custom: Configure your own focus and break durations.
- 🖥️ Smart HUD Overlay: A floating, always-on-top overlay widget automatically positioned in the bottom-right corner of your screen (safely above the taskbar) displaying remaining time and distraction state.
- 📊 Offline SQLite Database: Uses a WebAssembly-based SQLite database (
sql.js) to persist session logs, completion rates, and distraction events locally. - 🔒 Encrypted API Credentials: Encrypts your Gemini API key at rest using Electron's native
safeStorageAPI. - 🎨 Premium Dark Glassmorphic UI: Beautiful responsive layout with glowing neon indicators, custom glossy scrollbars, and dynamic inline vector SVG icons.
- Main Process (
src/main.ts): Controls window lifecycles, global tray setup, screen captures, native Windows balloon notifications, encrypted offline settings storage, and stateless API calls to Gemini. - Renderer Process (
src/renderer/renderer.ts): Handles DOM actions, tab navigation, timer counting state, speech recognition, audio playback of on-demand TTS audio chunks, and rendering analytics charts. - Database Layer (
src/database.ts): Initializes tables for settings, sessions, and distractions offline in SQLite WASM and runs automated migrations. - Preload Script (
src/preload.ts): Exposes IPC channels securely between the Electron main process and the front-end renderer.
For non-developers or quick setups, download the pre-compiled executables from the GitHub Releases page:
- Portable Version (
Multidoro.1.0.0.exe): Runs instantly without installation. - Installer Version (
Multidoro.Setup.1.0.0.exe): Installs the app to your system, adding it to your desktop and Start Menu.
- Node.js (v16+)
- A Gemini API Key (obtainable from Google AI Studio)
# Clone the repository
git clone https://github.com/msounhein/Multidoro.git
cd multidoro
# Install dependencies
npm installLaunch the application and paste your API key directly inside the Settings tab. The key will be encrypted and saved securely locally.
# Compile TypeScript and copy asset files
npm run build
# Start the application
npm startnpm run build: Compiles TS code and copies static renderer assets (HTML, CSS, templates, and WASM binary) to thedist/build output directory.npm start: Launches the compiled Electron binary.npm run dev: Optional watch script for compiling code during active development.
This project is licensed under the MIT License.