Eyezen

Cross-platform desktop eye care app — quiet, smart, out of your way

English | 简体中文

👀 Introduction

Eyezen is a lightweight desktop companion that protects your eyes without breaking your flow. It applies the 20-20-20 rule — every 20 minutes look ~6 m away for 20 seconds — and offers a Pomodoro alternative for deep-work sessions. Smart skipping during fullscreen apps, AFK periods, or whitelisted processes keeps reminders unobtrusive; built-in statistics and health analysis show whether the habit is actually sticking.

✨ Highlights

• ⏱️ Dual timer modes — 20-20-20 or Pomodoro, both fully configurable
• 🎯 Smart skipping — fullscreen detection, AFK idle, process whitelist, weekday schedule
• 📊 Statistics & Health Analysis — daily / weekly / monthly ECharts trends and an Eye-Care Index that quantifies the habit
• 🖥️ Multi-monitor reminders — break overlay on every connected display
• 🌓 Dark / Light + i18n — Simplified Chinese / English hot switching, native title-bar adaptation on Windows
• ⌨️ Global hotkeys & glassmorphic tray — start / skip / pause via shortcut; tray panel follows the icon, auto-hides on blur
• ⚡ Lightweight & tested — ~15 MB RAM via Rust + Tauri; 90%+ coverage on both frontend and backend, enforced in CI

📸 Screenshots

Core Experience

Statistics & Health Analysis

Pomodoro Mode

Configurable focus / short break / long break cycle alongside 20-20-20

🚀 Quick Start

Requirements: Node.js v18+, Rust (stable), platform-specific dependencies per Tauri v2 Prerequisites.

git clone https://github.com/rsecss/eye-zen.git
cd eye-zen
npm install
npm run tauri dev    # Development mode (hot reload)
npm run tauri build  # Production build → src-tauri/target/release/bundle/

Run all 8 local CI checks (fmt, clippy, cargo test, svelte-check, vitest, prettier, build, version sync) in one shot:

npm run ci

🛠️ Tech Stack

🏗️ Architecture

┌────────────────────────────────────────────────────────────┐
│  Frontend (Svelte 5)                                       │
│  Windows: main · tray-panel · tip-window · tip-minimal     │
│    invoke()  ─→ Tauri Commands (thin layer)                │
│    listen()  ←─ Typed Events (ts-rs bindings)              │
└────────────────────────────────────────────────────────────┘
                              │
┌────────────────────────────────────────────────────────────┐
│  Backend Services (Tauri State · 9 Arc-shared services)    │
│    Config · Timer · Detector · Window · Sound              │
│    Tray   · I18n  · Stat     · Hotkey                      │
│  Communication: watch channels + EffectSink trait          │
└────────────────────────────────────────────────────────────┘
                              │
┌────────────────────────────────────────────────────────────┐
│  Platform Layer — PlatformApi trait                        │
│  Windows · macOS · Linux X11/Wayland                       │
│  Capabilities: fullscreen detect · idle · foreground proc  │
└────────────────────────────────────────────────────────────┘

• Services are constructed once at startup and shared via Arc<AppServices> Tauri state; they communicate through typed tokio::sync::watch channels and an EffectSink trait so the timer state machine stays pure.
• Platform abstraction isolates OS-specific FFI behind the PlatformApi trait with per-capability degrade flags — the Settings UI greys out toggles whose capability is unavailable (e.g. AFK on Wayland).
• IPC contracts are auto-generated by ts-rs and centralized in src/lib/bindings/, eliminating a class of Rust↔TS drift bugs.
• Coding rules under .trellis/spec/ are the canonical source for layering, IPC, and platform conventions — read them before contributing.

⚙️ Configuration

Config is stored as config.toml in the system app data directory; the statistics database data.db lives alongside it.

All settings can be modified through the in-app Settings UI — you should never have to edit the TOML by hand.

🤝 Contributing

Contributions are welcome! Please read CONTRIBUTING.md for guidelines.

📄 License

GNU General Public License v3.0 or later