visionos-spatial-computing

Build visionOS spatial computing experiences with SwiftUI and RealityKit.

4|1|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/AutisticAF/claude-code-apple-dev-plugin --skill visionos-spatial-computing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: visionos-spatial-computing
Source: https://github.com/AutisticAF/claude-code-apple-dev-plugin/tree/main/skills/visionos-spatial-computing
Command: npx skills add https://github.com/AutisticAF/claude-code-apple-dev-plugin --skill visionos-spatial-computing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides clear, actionable guidance for building spatial computing experiences on Apple Vision Pro, removing the guesswork around windows, volumes, immersive spaces, RealityKit entity patterns, hand tracking, and spatial interactions so developers can deliver reliable 3D-enabled apps.

Core Features & Use Cases

  • Windows & Volumes: How to declare WindowGroup and volumetric windows, choose default sizes in meters, and open windows programmatically.
  • Immersive Spaces & Immersion Styles: Patterns for mixed, progressive, and full immersion, plus lifecycle handling when opening and dismissing immersive spaces.
  • RealityKit Integration: Loading USDZ/Reality assets, building entities in code, attachments for SwiftUI views, spatial audio, hover effects, and anchoring strategies (planes, hands).
  • Hand Tracking & Gestures: Implement spatial tap, 3D drag, two-hand rotation and magnify, gesture targeting, and ARKit hand-tracking providers with entitlement notes.
  • Scene Reconstruction & Best Practices: Plane detection, mesh reconstruction, collision/input requirements, and etiquette for shared space sizes.

Quick Start

Ask the skill to show how to create a volumetric WindowGroup with a RealityView that loads a USDZ model, adds CollisionComponent and InputTargetComponent, and attaches spatial tap and drag gestures while listing required entitlements.

Frequently Asked Questions about visionos-spatial-computing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a volumetric window in visionOS using SwiftUI?

To create a volumetric window in visionOS, declare a WindowGroup with a volumetric style in SwiftUI, set default sizes in meters, and use a RealityView to load USDZ models and manage 3D content within the defined space.

How does hand tracking work on visionOS and what entitlements are required?

Hand tracking on visionOS uses ARKit hand-tracking providers to capture spatial data. Developers must request and verify specific ARKit hand-tracking entitlements within the app's configuration to authorize access to skeletal joint positions.

What is the difference between mixed, progressive, and full immersion styles in visionOS?

Immersion styles in visionOS control how much a user sees the real world: mixed overlays 3D content on reality, progressive blends based on user dial input, and full replaces the environment entirely, each requiring specific lifecycle handling.

How do I add spatial gestures and collision targeting to RealityKit entities?

To add spatial gestures to RealityKit entities, attach CollisionComponent and InputTargetComponent to the entity, then bind spatial tap and 3D drag gestures in SwiftUI to detect user interactions accurately.

Can I load USDZ assets and implement spatial audio in a visionOS RealityView?

Yes, you can load USDZ or Reality assets into a visionOS RealityView and attach spatial audio components directly to entities, enabling distance-based 3D soundscapes that anchor to your models.

When do I need scene reconstruction and plane detection in visionOS apps?

Scene reconstruction and plane detection are needed when your visionOS app requires world anchoring, collision with real surfaces, or mesh generation to allow virtual entities to interact physically with the user's environment.