What problem does it solve?
Building a working Isaac Sim scene requires coordinating asset import, physics configuration, robot control, rendering, and validation across many APIs; this Skill decomposes a natural-language request into verified capabilities and routes each to the right specialist skill so the final simulation actually runs and renders correctly.
Core Features & Use Cases
- Task decomposition and routing: Maps requests to specialist skills (usd-pipeline, isaac-sim-rendering, physics-simulation, isaac-sim-validator) and verifies each capability in isolation before integration.
- Multi-robot fleet simulation: Provides grid placement, collision groups, navigation kinematics, and VRAM-based scaling limits for warehouse scenes with Nova Carter, Spot, and other robots.
- Rendering and debugging playbooks: Includes RT2/PathTracing settings, ACES tonemapping, lighting defaults, and symptom-to-fix tables for black frames, OOM crashes, and articulation issues.
- Use Case: Ask for "6 Nova Carter robots in a warehouse with 100 racks, captured on video" and the Skill composes the USD scene, configures the PhysicsScene, launches a headless sim, validates the render, and delivers the video.
Quick Start
Ask the agent to create an Isaac Sim warehouse scene with several Nova Carter robots, run it headless on your GPU, and deliver a validated rendered video.