daedalOS

daedalOS is a desktop environment in the browser. Its core system components include:

• File System: Powered by BrowserFS, it supports drag & drop, ZIP/ISO read support, and writes to IndexedDB. It includes a File Explorer with thumbnail/details views and context menus for file manipulation (cut, copy, rename, etc.).
• Windows: Resizable and draggable windows that persist size, position, and maximized states. They feature minimize, maximize, and close functionality with animations.
• Start Menu: An expandable sidebar containing an apps list, document/media shortcuts, and a Power option to clear the session. It can be opened with SHIFT+ESC (or the Windows Key in fullscreen).
• Taskbar: Includes window hover previews (Peek), a focused window indicator, a search menu with recent files, and an AI Chat Agent.
• Clock: Runs in a Web Worker using OffscreenCanvas, supporting NTP server time mode and syncing to the system clock.
• Background & Screensaver: Supports dynamic animated wallpapers (Waves, Hexells, Matrix, etc.), custom images/videos, and various API-driven sources like NASA's Astronomy Picture of the Day.

daedalOS is a browser-based desktop environment built using a wide array of open-source frameworks, core libraries, and specialized system/app libraries. Developers looking to extend or understand the system should be aware of the following key technologies used:

Core Frameworks

• UI & Logic: React, Next.js, TypeScript, styled-components
• Animation & Interaction: Framer Motion, react-rnd
• File System: BrowserFS

Specialized Capabilities

• Media & Graphics: FFmpeg (via ffmpeg.wasm), ImageMagick (WASM), Video.js, PDF.js, and various image decoders (jxl.js, libheif-js, UTIF.js).
• Emulation & Virtualization: v86 (Virtual x86), emulatorjs, js-dos, Ruffle (Flash), and Pyodide (Python).
• Code & Text Editing: Monaco Editor, Xterm.js, Vim.js, and TinyMCE.
• AI & Machine Learning: WebLLM and Web Stable Diffusion.
• Networking & Data: isomorphic-git, Lunr (search), and various CORS proxies for service integration.

External Services

• The system integrates with several public APIs including NASA (APOD), The Met (MMoA), Art Institute of Chicago, and various DNS-over-HTTPS providers (Cloudflare, Google).

Tracky Mouse is a JavaScript library that allows you to implement head-tracking UI in web applications. You can install it via npm and then initialize it by setting the dependenciesRoot and calling loadDependencies() followed by init().

Note: The library uses clmtrackr.js for lightweight tracking and switches to facemesh (TensorFlow.js) for higher accuracy once loaded. You must manually handle the mapping of the tracked coordinates to DOM events (like PointerEvent) to make the tracking interact with your application's UI.

<script src="node_modules/tracky-mouse/tracky-mouse.js"></script>
<script>
	TrackyMouse.dependenciesRoot = "node_modules/tracky-mouse";
	TrackyMouse.loadDependencies().then(function() {
		TrackyMouse.init();

		// Example: Mapping TrackyMouse output to PointerEvents
		TrackyMouse.onPointerMove = (x, y) => {
			const target = document.elementFromPoint(x, y) || document.body;
			// ... dispatch PointerEvent('pointermove', ...) to target
		};
	});
</script>