LLM Limits

A floating desktop monitor for AI quota usage (Codex/OpenAI, Claude, Gemini), built with Electron + React + TypeScript.

Features

• Always-on-top compact monitor window
• Session/period quota toggle in the monitor
• Reset timestamp display with configurable:
  • 12H / 24H time
  • DD.MM.YYYY / MM.DD.YYYY date order
  • . or / date separator
• Settings auto-open on first run (or when provider config is missing)
• Optional auto-start on OS login
• Window position memory + snap-to-edge placements
• Auto period-window warmup: sends a minimal query when the Codex period window expires to establish a new window immediately (API key and OAuth/CLI supported)
• CLI path checker: test button in Settings to verify codex and gcloud are available in PATH
• Active selection highlighted with a visible ring in Settings toggle groups

Tech Stack

• Electron
• React + TypeScript
• Tailwind CSS
• electron-store (local settings persistence)

Prerequisites

• Node.js 20+
• npm
• Windows recommended for packaging (.exe)

Install

npm install

Run in Development

npm run dev

Build

npm run build

Build output is generated in out/.

Create Windows Installer (.exe)

npm run dist

Installer/artifacts are generated in dist/ (for example NSIS installer and win-unpacked).

Configuration

Open Settings from the app context menu/tray.

Available settings include:

• Provider credentials/configuration
  • Codex/OpenAI
  • Claude (API or web-login integration)
  • Gemini
• Update frequency (minutes)
• Time format (24H/12H)
• Date format + separator
• Auto-start on OS login
• Debug logging

Contributing

Contributions are welcome.

1. Fork the repository
2. Create a feature branch
3. Make your changes with clear commit messages
4. Open a pull request with a short description and test notes

Please use Issues for bug reports and feature requests.