agency-macos-spatial-metal-engineer

Render large dynamic graph datasets with instanced Metal for stereoscopic 90fps playback.

Updated Feb 11, 2026
One-click install
npx skills add https://github.com/augustoheiss/LogicDefense --skill agency-macos-spatial-metal-engineer-augustoheiss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agency-macos-spatial-metal-engineer
Source: https://github.com/augustoheiss/LogicDefense/tree/main/.gemini/skills/agency-macos-spatial-metal-engineer
Command: npx skills add https://github.com/augustoheiss/LogicDefense --skill agency-macos-spatial-metal-engineer-augustoheiss

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a specialist agent that eliminates the complexity of designing and optimizing native Swift and Metal renderers for massive dynamic graph datasets and streaming stereoscopic output to Vision Pro, ensuring high frame rates and responsive spatial interactions.

Core Features & Use Cases

  • GPU-scalable rendering: Instanced Metal pipelines, triple buffering, frustum culling, LOD, and aggressive draw-call batching to sustain tens of thousands of nodes at target frame rates.
  • Vision Pro integration: Compositor Services and RemoteImmersiveSpace streaming, gaze and pinch gesture handling, raycast-based selection, and spatial transition management for immersive code-graph visualization and collaborative review.
  • Use Case: Build a macOS companion renderer that streams an interactive 3D code-graph to Vision Pro for immersive debugging, allowing engineers to explore and manipulate 25k+ nodes in real time.

Quick Start

Design a Metal-based renderer that streams stereo frames to Vision Pro and maintains 90fps with 25k nodes.

Frequently Asked Questions about agency-macos-spatial-metal-engineer

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

FAQPage Schema
How do I render large dynamic graph datasets in Metal while maintaining 90fps stereoscopic playback on Vision Pro?

To render large dynamic graph datasets in Metal at 90fps stereoscopic, apply instanced rendering pipelines, aggressive draw-call batching, frustum culling, LOD, and triple buffering to sustain 25k+ nodes under 1GB GPU memory.

What is the best way to stream stereo frames from a macOS companion renderer to Vision Pro?

The best way to stream stereo frames to Vision Pro is using Compositor Services and RemoteImmersiveSpace workflows, enabling spatial transitions and immersive code-graph visualization directly from the macOS companion renderer.

How does gaze-based selection work for spatial interactions in a Metal renderer?

Gaze-based selection in a Metal spatial renderer works by handling gaze and pinch gestures through raycast-based selection, allowing engineers to interactively explore and manipulate nodes in real time.

Can I sustain 90fps with 25k nodes in a Swift Metal renderer without exceeding 1GB GPU memory?

Yes, you can sustain 90fps with 25k nodes under 1GB GPU memory by applying instanced Metal pipelines, frustum culling, triple buffering, and aggressive draw-call batching validated via Metal System Trace profiling.

Why does my Metal spatial renderer drop frames when rendering massive dynamic graphs?

A Metal spatial renderer drops frames when rendering massive dynamic graphs due to unbatched draw calls and missing frustum culling, requiring instanced rendering and triple buffering to maintain stereoscopic 90fps.

Do I need Compositor Services to build a macOS companion renderer for Vision Pro?

Yes, Compositor Services are needed to build a macOS companion renderer for Vision Pro, providing the framework for streaming stereo frames, spatial transitions, and immersive code-graph visualization.