A specialized development skill for implementing and debugging ARKit-powered spatial features on visionOS using ARKitSession and RealityKit.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install arkit-visionos-developer
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install arkit-visionos-developer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The ARKit visionOS Developer skill is a comprehensive technical resource designed to streamline the creation of immersive spatial experiences. It focuses on the specialized architectural requirements of visionOS, where ARKit serves as the data layer that feeds spatial information into RealityKit for rendering. By leveraging this skill within Openclaw Skills, developers can efficiently manage the lifecycle of ARKit sessions, handle complex privacy authorizations, and bridge real-world data into virtual environments.
This skill bridges the gap between raw hardware data and high-level scene reconstruction. It provides developers with the patterns necessary to handle world tracking, hand gestures, and environmental mesh data without blocking the main actor, ensuring fluid performance in high-stakes spatial computing applications.
To integrate this skill into your workflow, ensure your project is configured for visionOS development and update your Info.plist with the required usage descriptions:
# Required Info.plist keys for ARKit visionOS development
NSWorldSensingUsageDescription "Used for world tracking and plane detection"
NSHandsTrackingUsageDescription "Used for hand gesture interaction"
NSMainCameraUsageDescription "Used for camera-based tracking features"
Ensure you are using a Full Immersive Space for most providers and maintain a strong reference to your ARKitSession within your data model to prevent premature session termination.
The skill manages spatial data through a structured provider-anchor system. Below is the primary taxonomy for Openclaw Skills implementation:
| Data Component | Responsibility | Mapping Pattern |
|---|---|---|
| ARKitSession | Lifecycle & Auth | Single instance per ImmersiveSpace |
| Data Provider | Sensor Data Input | Specialized streams (Hand, Plane, World) |
| Anchor Updates | Spatial Changes | Reconcile Added/Updated/Removed states |
| State Model | App Logic | Normalized IDs mapped to RealityKit Entities |
| RealityView | Presentation | Rendering layer bridged via ARKitAnchorComponent |
Loading
A professional-grade deepfake and face-swap detection utility that analyzes media files for AI manipulation using the Scam.ai API.

Send and manage multi-platform push notifications seamlessly using Openclaw Skills and the Membrane framework.

A powerful integration for programmatically capturing website screenshots and converting web pages to high-quality PDFs.

A lightweight utility that allows AI agents to fetch and display the current system time in a user-friendly format.

A local, high-performance transcription skill that integrates OpenAI Whisper and ffmpeg into the OpenClaw Gateway host.

A mathematical position sizing tool for cryptocurrency traders based on the Kelly Criterion to maximize long-term growth while minimizing ruin risk.








































