Make your time count.
An AI-native desktop productivity app that turns your tasks into time records —
so you always know where your day actually went.
Most productivity apps help you write tasks. Yolo helps you do them — and understand how long they truly take.
The real problem isn't making lists. It's not knowing when you started, how long something took, or what you actually accomplished at the end of the day. Yolo keeps a precise time record, so you can review, reflect, and improve.
Plan the day. Run one focus. Review the truth.
- Create tasks with category, priority, estimated time, and due date
- Full lifecycle: start → pause → resume → stop → complete (or drop)
- One active task at a time — eliminate multitasking noise
- Automatic per-session timer that survives app restarts
- Stop-session notes: what you did, blockers, next steps, completion rate
- Accurate cross-midnight time entry splitting
- Today view — task list + live timer + time log + daily summary
- Compare estimated vs. actual time per task
- Category totals and daily focus time at a glance
- 7-day history with per-day breakdowns
- Plan recurring work with start and end times
- Carry overdue work forward without losing the original context
- Move tasks between today, tomorrow, next week, and backlog
- See your life in weeks, mapped against real tracked focus
- Spot high-focus weeks and drill back into the underlying days
- Use long-horizon perspective without leaving the daily workflow
- A tool-calling chat agent that reads your day and can create, update, schedule, start, complete, or drop tasks through validated tools
- Choose autonomy in Settings → AI: Plan/Ask queues write cards for approval; Auto applies reversible writes and keeps Revert available
- History-aware: learns from your real time records — estimate-vs-actual calibration, slip & blocker detection, and an automatic weekly review
- Grounds advice in your actual numbers and applies your real estimation ratios when planning the day
- Numbers are computed deterministically; the model only explains them — so it never makes them up
- Bring your own AI key (OpenAI / Anthropic / Gemini / compatible) in Settings → AI
- Ships an MCP server that exposes your tasks and time records to AI agents
- Works with Claude Desktop, Claude Code, Cursor, and any MCP-compatible tool
- Ask "where did my time go today?" directly from your AI assistant
- External agents can read and manage tasks; set
YOLO_MCP_READONLY=1when you want a read-only server
| Layer | Technology |
|---|---|
| 🖥 Desktop Shell | Tauri v2 (Rust) |
| ⚛️ UI Framework | React 18 + TypeScript + Vite |
| 🎨 Styling | Tailwind CSS + Radix UI |
| 🐻 State Management | Zustand |
| 🗄 Database | SQLite via tauri-plugin-sql |
| 🎞 Animation | Framer Motion |
| 🔧 Utilities | date-fns, lucide-react |
| 🤖 AI Integration | MCP (Model Context Protocol) |
- Node.js ≥ 18
- Rust (for Tauri)
- Tauri CLI prerequisites for your OS
# Clone the repo
git clone https://github.com/your-username/yolo.git
cd yolo
# Install dependencies
npm install# Run the web frontend only (browser, no Tauri)
npm run dev
# Run the full desktop app (recommended)
npm run tauri dev# Build the frontend bundle
npm run build
# Build the native desktop app (distributable)
npm run tauri build# Run all tests once
npm test
# Run in watch mode
npm run test:watchYolo stores its task, time-entry, category, and settings records in SQLite (yolo.db) through Tauri's SQL plugin.
React components never touch SQL directly — all data access goes through repository modules in src/db/.
Schema:
| Table | Purpose |
|---|---|
tasks |
Task definitions, status, estimates |
time_entries |
Per-session time records |
categories |
User-defined task categories |
settings |
App preferences and theme |
Yolo ships a standalone MCP server in mcp/ that exposes your tasks and time records to any MCP-compatible AI agent.
cd mcp && npm install && npm run buildAvailable tools:
| Tool | Description |
|---|---|
list_tasks |
List tasks with optional filters |
get_task |
Get a single task by ID |
list_time_entries |
List time records, filter by date or task |
daily_summary |
Get a summary of a specific day |
list_categories |
List all categories |
See mcp/README.md for the full Claude Desktop config and integration details.
The core desktop workflow is complete. The next phase makes Yolo truly AI-native.
- 🧠 AI assistant — in-app tool-calling agent with autonomy levels, undo, and history-aware planning
- 🔌 MCP write tools — external agents can create and update tasks; read-only mode remains available
- 🔮 Proactive coaching — unprompted morning plans, focus-rhythm placement, goal-driven planning
- 💬 AI-native capture — natural-language quick-add and smart stop-notes
- 🖥 Desktop power features — system tray, global shortcuts, mini floating timer, notifications
See docs/ROADMAP.md for the full phased plan and architecture decisions.
yolo/
├── src/ # React frontend
│ ├── components/ # UI components (layout, pages, primitives)
│ ├── db/ # Repository layer (all SQL access)
│ ├── hooks/ # Custom React hooks
│ ├── services/ # AI assistant + retrospective analytics
│ ├── stores/ # Zustand state stores
│ ├── types/ # Shared TypeScript types
│ └── utils/ # Helper utilities
├── src-tauri/ # Tauri (Rust) backend
│ ├── src/ # Rust source
│ └── icons/ # App icons (all sizes)
├── mcp/ # MCP server for AI agent access
└── docs/ # Documentation and roadmap
Contributions are welcome! Please open an issue to discuss what you'd like to change before submitting a pull request.
- Fork the repository
- Create your feature branch:
git checkout -b feat/amazing-feature - Commit your changes:
git commit -m 'feat: add amazing feature' - Push to the branch:
git push origin feat/amazing-feature - Open a Pull Request
This project is private. All rights reserved.