skills.rest
NVIDIA-Omniverse
NVIDIA-OmniverseOFFICIAL
Explore all skills by this author →
Website𝕏@nvidiaomniverseGitHub

tune-parameters

Guide iterative parameter tuning for USD scene-optimizer operations toward quality targets.

31|4|Updated May 21, 2026
Search Tags:#parameter tuning#guide authoring#usd optimization#scene optimizer#visual diagnosis#usdview inspection#pxr inspection
One-click install
npx skills add https://github.com/NVIDIA-Omniverse/usd-optimize --skill tune-parameters
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tune-parameters
Source: https://github.com/NVIDIA-Omniverse/usd-optimize/tree/main/.agents/skills/tune-parameters
Command: npx skills add https://github.com/NVIDIA-Omniverse/usd-optimize --skill tune-parameters

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Tune-parameters helps you iteratively find the right parameter values for a specific NVIDIA USD scene optimization operation so your USD results meet visual and structural goals without guessing.

Core Features & Use Cases

  • Interactive, screenshot-driven tuning loop: inspect input/output visually and adjust one parameter at a time toward an acceptance target.
  • Tiered operation knowledge loading: uses a full per-operation guide when available, otherwise falls back to session logs or C++ source parameter definitions.
  • USD-aware context gathering: auto-inspects USD stage metrics (units, bbox, prims) and optionally generates rendered previews or usdview-based inspection for deeper diagnosis.
  • Guide authoring mode: supports developers creating new .agents/operations/<key>.md tuning guides by extracting argument metadata from C++ source and applying a guide template.

Quick Start

Ask the AI to tune-parameters for operation X using your input USD path and tell it your goal (for example, reduce polycount or close holes), optionally providing an output screenshot for comparison.

Frequently Asked Questions about tune-parameters

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

FAQPage Schema
How do I tune USD optimization parameters for a specific scene operation?▼

To tune USD optimization parameters, iteratively adjust one value at a time toward an acceptance target using visual diagnosis and stage inspection to reach your defined quality goal.

What is the best way to find sensible starting configurations for USD scene optimization?▼

Finding sensible starting configurations for USD scene optimization relies on tiered knowledge loading, which uses per-operation guides, session logs, or C++ source parameter definitions as references.

How does visual diagnosis work during USD parameter tuning?▼

Visual diagnosis during USD parameter tuning works by auto-inspecting stage metrics like units, bbox, and prims, optionally generating rendered previews or usdview-assisted inspection for comparison.

Can I inspect USD stage metrics with pxr if I do not have a full operation guide?▼

You can inspect USD stage metrics with pxr through best-effort fallback, which extracts parameter argument metadata directly from C++ source definitions when a full per-operation guide is unavailable.

How do I author a new tuning guide for custom USD optimization operations?▼

Authoring a new tuning guide for USD optimization operations uses a guide authoring mode that extracts argument metadata from C++ source and applies a structured template for `.agents/operations`.

Related Skills

Explore More Software Engineering Skills→
NVIDIA-OmniverseNVIDIA-Omniverse
201

cloning-prims

Clone USD prim subtrees to new paths on a runtime stage.

Official
Intermediate
jakildevjakildev
8

render-debug-loop

Automate build, run, screenshot, and visual regression evaluation for the Irreden Engine.

Community
Advanced
NVIDIA-OmniverseNVIDIA-Omniverse
31

inspect-asset

Inspect USD stage files and generate structured JSON inspection reports.

Official
Basic
PubCyBerryPubCyBerry

loading-usd

Load and compose USD scene content for renderer applications.

Community
Advanced