Omniverse Usd Performance Tuning

OpenAIOfficialCreated by NVIDIA OmniverseDistributed by OpenAI

Skill
Workflow
Software development
Apache-2.0
Medium risk
Has scripts

From the author

Top-level workflow skill for USD performance diagnosis and optimization. Use for slow loading, high memory, low FPS, or 'optimize my scene' requests; delegates auth/runtime setup to Phase 0 owners.

— the description by NVIDIA Omniverse

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

The omniverse-usd-performance-tuning skill provides a structured, end-to-end procedure for addressing performance bottlenecks in USD assets. It manages the lifecycle of performance tuning, from initial runtime context setup and baseline profiling to composition auditing, validation, and targeted scene optimization. The skill enforces a rigorous milestone-based approach, ensuring that all destructive operations are planned and approved, and that performance improvements are verified through comparative profiling and standardized reporting.

What it does

Orchestrates the end-to-end USD performance tuning lifecycle.
Validates runtime environments and handles authentication for remote assets.
Conducts composition audits and structural assessments of USD stages.
Executes validated optimization operations using Scene Optimizer.
Generates structured optimization reports comparing baseline and post-optimization metrics.
Enforces safety gates for destructive operations and mutation.

When to use it

When a USD scene exhibits slow loading times or low frame rates.
When memory usage is excessively high or GPU crashes occur.
When you need to optimize a scene for better interactivity or performance.
When you need to diagnose CAD conversion quality issues.
When you need to validate a scene before and after applying optimizations.

Advantages

Ensures consistent performance diagnosis through a standardized workflow.
Prevents accidental data loss by requiring approval for destructive operations.
Provides clear, evidence-based reporting on performance improvements.
Integrates seamlessly with existing Omniverse validation and optimization tools.
Maintains strict separation between diagnosis, planning, and execution phases.

Usage examples

“This USD scene loads very slowly; please triage what I should check first.”: The agent will initiate the runtime context gate, perform a composition audit, and provide a structured plan to identify the source of the slow load time.
“Optimize my CAD scene for better performance.”: The agent will verify the runtime, assess the CAD conversion quality, and propose a milestone-based optimization plan including validation and profiling.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/openai/omniverse-usd-performance-tuning/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip omniverse-usd-performance-tuning.zip -d .agents/skills

Documentation by NVIDIA Omniverse

Omniverse USD Performance Tuning

When to Use

Use this workflow for broad performance asks such as slow loading, high memory, low FPS, GPU crashes, conversion-quality triage, or generic requests to optimize a USD scene.

Instructions

Start from the mandatory runtime context gate before producing tuning output, unless the prompt is only asking for a static classification test.
Classify broad optimization requests as ready_to_plan; reserve approval_required for prompts that explicitly name a destructive operation to execute before planning.
Plan the full canonical chain through optimization-report, preserving the structured milestone order and the profile-stage:baseline / profile-stage:after labels when listing milestones.
Invoke downstream skill bodies only when their phase is reached, and keep raw runtime artifacts on disk while reading compact summaries.

Frontmatter keeps version and tools at top level for agentskills.io runtime compatibility. NVCARPS discoverability fields live under metadata.

Output Format

Return a plan or status summary that names the selected entry skill, uses ready_to_plan for generic optimization requests, includes the full milestone chain through optimization-report, and labels profile phases as profile-stage:baseline and profile-stage:after. For structured outputs, the broad-optimization milestone subsequence is omniverse-usd-performance-tuning -> profile-stage:baseline -> usd-structure-assessment -> usd-validation-runner -> restructure-decision -> apply-restructure -> so-run-validators -> so-interpret-validators -> so-run-operations -> profile-stage:after -> compare-profiles -> optimization-report. End-to-end execution should produce an optimized stage when mutation runs and a report conforming to the optimization-report reference's schema (scripts/optimization-report.schema.json within that reference).

Use this workflow for broad performance asks such as slow loading, low FPS, high memory, GPU crashes, conversion quality, or "optimize my scene."

Entry skill rule

This skill is the named entry point for broad performance work whenever the agent has any verified way to do that work. Runtime probing details live in setup-usd-performance-tuning; this rule only decides which skill owns the user-facing performance request.

If the setup probe shows any verified runtime path - Kit, standalone, or even a partial stack such as Asset Validator only - enter here. If the user's requested tool is missing, return the specific blocked_code (blocked_missing_scene_optimizer, blocked_missing_so_operation, etc.) instead of substituting another workflow.
Enter at setup-usd-performance-tuning only when no runtime path is verified and runtime choice/setup is the first unresolved problem.
For omniverse:// assets, enter at omniverse-authentication first. Authentication precedes setup and triage for remote assets.

The decision is about ownership, not order. Setup, authentication, and triage all run in their normal phase order; this rule only fixes which skill the agent names as the entry skill in its response.

Runtime context — session-start gate (mandatory)

Before any other tuning output, follow the mandatory session-start gate in skills/omniverse-usd-performance-tuning/references/setup-usd-performance-tuning/references/runtime-context-header.md. That reference owns output_path, the canonical setup-preflight.json location, Format A/Format B, and the "do not improvise a silent probe" anti-pattern.

Required outcomes:

Missing or unreadable preflight: invoke setup-usd-performance-tuning.
Present preflight: print Format A and wait for the user to choose Continue, Change Kit, Switch to standalone, or Re-run probe.
Confirmed runtime in the same session: use compact Format B for follow-up status.

[Kit: {runtime_context.kit.application} {runtime_context.kit.version}  |  SO: {runtime_context.sceneOptimizer.version}  |  AV: {runtime_context.assetValidator.version}]

Runtime artifact token budget

Before reading Kit logs, Asset Validator CSVs, Scene Optimizer logs, Tracy CSVs, or other runtime output, follow references/runtime-artifact-token-budget.md. Keep raw artifacts on disk, read summary JSON first, and use bounded log snapshots instead of full dumps or live streams.

Plan-time vs execution-time approval

approval_required at planning time is reserved for requests that explicitly name a destructive operation. Use the following rule when deciding between ready_to_plan and approval_required:

approval_required at planning time — the user's request itself names a destructive operation: "flatten this stage", "decimate the meshes", "merge prototypes", "delete unused prims", or any specific named mutation that cannot be undone within the same workflow. In this case the agent's first response must be an approval prompt that names the operation, before the agent commits to a plan that executes it.
ready_to_plan at planning time — the user's request is general: "optimize this scene", "make it load faster", "reduce GPU memory", "improve interactivity". The agent lays out the full plan, including any destructive operations the plan would invoke (for example so-run-operations with mergeMaterials), without withholding the plan itself. Approval for each destructive operation is requested alongside plan approval.

The distinction is between authorising a plan and authorising a destructive action. A general optimisation request authorises planning; it does not authorise execution of specific destructive operations.

For structured runtime-test responses and similar planning summaries:

A future restructure-decision prompt is a planned user-decision gate, not a reason to set the top-level response decision to approval_required for a generic optimization request.
For a generic optimization request, set decision: "ready_to_plan" and include the full intended chain in both committed_milestones and planned_phases, through optimization-report.
It is valid for gates_observed to include asks_user_for_restructure_decision while the top-level decision remains ready_to_plan.
Whenever a chain names profile phases, use the exact labels profile-stage:baseline and profile-stage:after; do not emit the ambiguous bare profile-stage token.
Start structured milestone lists with omniverse-usd-performance-tuning as the owning entry skill. Include setup-usd-performance-tuning only as additional Phase 0 context, not as a replacement for the entry skill milestone.
For broad optimization requests, preserve the milestone subsequence from Output Format above exactly, with optional extra analysis steps inserted only where they do not reorder it.
Do not list so-run-validators or so-interpret-validators before restructure-decision in broad optimization milestone summaries. Phase-aware validator routing still happens through usd-validation-runner; the SO validator executor/interpreter milestones appear after the restructure decision path in the structured plan contract.

Output expectation

End-to-end optimization work should produce both an optimized USD stage, when mutation is executed, and a structured optimization report conforming to the optimization-report reference's scripts/optimization-report.schema.json. The HTML report must be rendered from references/report-templates/optimization-report.html.template via render_preview.py — never hand-write HTML. Diagnosis-only work should still end with a report or summary that states no optimized stage was written.

Purpose

Route digital twin USD performance requests into the right diagnostic and optimization workflow while preserving evidence before mutation.

Prerequisites

Stage path or enough context to identify the target asset.
User goal: diagnosis only, validation, profiling, or processor execution.
Runtime availability status from setup-usd-performance-tuning when not already known.
Permission status for in-place mutation vs writing a separate optimized output.

Examples

"This USD loads slowly; triage what to check first."
"Route a low-FPS CAD scene through the performance workflow."

Triage order

Runtime gate. Follow the mandatory session-start gate above before validation, profiling, or optimization. Do not scan, probe, install, or pick Kit/standalone runtimes directly in this skill; setup-usd-performance-tuning owns probe/chooser/install dispatch and writes the preflight consumed here.
Identify the target problem:
- Load time.
- FPS or interactivity.
- GPU or system memory.
- Crash or device lost.
- CAD conversion quality.
- Validation failure.
Gather minimum context:
- Stage path and size.
- Whether the stage is local, mounted, or omniverse:// remote. For remote assets, route through omniverse-authentication before first open.
- Kit or USD runtime.
- Whether the workload is CAD, VFI, AIF, Isaac, or generic OpenUSD.
- Whether in-place mutation is allowed.
- Whether the user wants diagnosis only or processor execution.
Route:
- USD composition questions: usd-structure-assessment (composition is now part of the SA umbrella; deeper detail in skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/composition-audit.md).
- Validation and content issues: usd-validation-runner (master router; routes to validate-* family or so-run-validators based on intent).
- Edit/output decisions: usd-edit-target-planner (also owns variant/payload gates).
- Repeated copied hierarchy or high mesh count with no instancing: usd-hierarchy-dedupe-candidates.
- Restructure decision (monolithic stage, asset boundary materialization): restructure-decision.
- CAD converter settings: read references/cad-conversion/README.md (niche pre-USD concern; see reference for details).
- Scene Optimizer: so-run-validators, so-interpret-validators, so-run-operations.

Optimization ordering

Follow the canonical ordering in workflow.md § Operation ordering invariants. The high-level rule: prototypes first → per-asset validation → stage-level operations last. The workflow reference owns the full invariant list (meshCleanup before decimateMeshes, deduplication before decimation, never merge if instanced, etc.) and the analysis-only ops catalogue.

Rules

Always run composition audit before mutation.
Always validate before and after processor execution.
Optimize prototypes before per-asset validation.
Do not run whole-stage mesh deduplication on very large CAD scenes before checking for hierarchy-level reuse.
Do not recommend a fixed optimization stack without bottleneck evidence.
Do not invent numeric thresholds or expected percentage wins.
Prefer canonical SO ops over specialty / documentary ones. The op curation in references/operations/_curation.json classifies every op as canonical, specialty, analysis, documentary, or deprecated. When more than one op could resolve the same finding, recommend the canonical one first and only reach for a specialty op when the user explicitly asks or the rationale warrants it. Specifically:
- For vertex welding, prefer canonical meshCleanup with explicit flags over the standalone mergeVertices op. The standalone op is a legacy/specialty surface; use upstream usd-optimize for the operation mechanics and local approval policy before mutating.
- For hierarchy dedupe, recommend usd-hierarchy-dedupe-candidates + apply-restructure (the USD-authored rewrite path).
- For per-mesh dedupe, recommend deduplicateGeometry (canonical) over findCoincidingGeometry (analysis — produces a report, not a change).
- Do not recommend documentary-status ops (e.g., boxClip, deletePrims, removeAttributes, removeUntypedPrims, merge outside its narrow non-instanced case) without an explicit user request. Documentary ops survive in the per-op references/operations/<key>.md routing stubs for completeness but are excluded from agent-initiated recommendations.
- Specialty ≠ documentary. Ops classified as specialty in _curation.json either (a) have validator-finding evidence that wires them into the so-interpret-validators chain (e.g. sparseMeshes, optimizePrimvars), or (b) are load-bearing escape hatches needed for specific downstream contexts (e.g. primitivesToMeshes when output must be UsdGeomMesh, utilityFunction for instancing toggles and material rebinding, pythonScript for so-create-proxy recipes). Recommend specialty ops when their validator fires OR when their downstream context applies — the suppression above only targets documentary ops.

Limitations

Does not replace downstream reference instructions; load each required reference before executing it.
Does not install runtimes directly; follow setup or install references when requirements are missing.
Does not authorize mutation when the user has not allowed writes.

Troubleshooting

If runtime status is unclear, run setup-usd-performance-tuning before profiling or validation.
If the reported problem is vague, gather stage path, workload type, and whether diagnosis or execution is requested.
If the workflow suggests mutation before evidence, return to baseline profiling and composition audit first.

References

Before routing, read:

skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/optimization-tradeoffs.md — identify which pipeline phase the scene is in (extraction, structuring, or optimization). The right action depends on the phase.
skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/factory-level-structuring.md — understand the three pillars (assets, aggregation, animation) and the seven-step structuring pattern.

If you have network access, prefer the live URLs (noted in each reference file) for the most current version.

Required execution flow

Read references/workflow.md for the canonical Phase 0-7 flow, including Kit/standalone branches, validator-stack routing, operation ordering, termination conditions, duration hints, and the optional iteration pattern. The compact root map at references/skill-map.md only routes agents into this workflow.

Do not treat downstream phase names as plain checklist labels. Before executing each step, load that phase's nested README.md reference and follow its instructions. Claude Code only exposes the public catalog skill; it does not recursively inject profile-stage, usd-structure-assessment, or other nested references.

The final deliverable must come from optimization-report: save both the structured JSON report and the generated Markdown summary. Do not substitute an ad hoc SUMMARY.md or chat-only recap for the optimization report.

For deeper subtopic guidance, consult the references:

skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/composition-audit.md, skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/layer-health.md - subtopic detail for SA's Phase 1 checklist.
skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/instancing-readiness/references/instancing-tradeoffs.md - merge safety, decision tree for instancing choices.
skills/omniverse-usd-performance-tuning/references/usd-structure-assessment/references/usd-edit-target-planner/references/variants-payloads.md - deeper variant/payload trade-offs (gates are inline in usd-edit-target-planner).
references/cad-conversion/README.md - CAD converter settings.
references/upstreams/usd-optimize.md - upstream SO mechanics and prebuilt package resolution.
skills/omniverse-usd-performance-tuning/references/usd-validation-runner/references/so-run-validators/references/infrastructure.md - local handoff for SO validator infrastructure.
skills/omniverse-usd-performance-tuning/references/usd-validation-runner/references/validation-scoping.md - tier 1/2/3 plan + scene-aware adjustment.
skills/omniverse-usd-performance-tuning/references/optimization-report/references/optimization-report-template.md - the data contract every phase populates.

For full Kit runtime profiling (FPS, frame time, Hydra/RTX metrics), refer to the external profiling skills at NVIDIA/omniperf.

Files

references/optimization-report/README.md

Optimization Report

When to Use

Use this reference only at the end of the workflow after profile comparison and validation evidence are available.

Instructions

Confirm the target asset, artifact, or user intent and check the prerequisites listed below.
Read only the referenced files needed for the current phase, failure mode, or output contract.
Follow the workflow, rules, and safety gates in this reference before invoking downstream references or shell commands.
Return the result using the Output Format section and name any blocked prerequisite or unresolved user decision.

Use this reference as the final step in the optimization flow — after compare-profiles has produced its verdict. This reference assembles the complete optimization story into a structured report.

Pre-flight Checklist

Before generating the report, re-read and confirm:

[ ] HTML must use the committed renderer — invoke references/report-templates/render_preview.py with --fixture and --output. Never hand-write HTML. Never use LLM-generated HTML.
[ ] JSON conforms to schema — read scripts/optimization-report.schema.json in full before writing. Do not guess field names from memory.
[ ] Runtime context copied verbatim from setup-preflight.json.
[ ] Score computed deterministically — weighted average of metric groups, not hand-assigned.

Anti-pattern: hand-written HTML

If find or glob does not locate the template, you are looking in the wrong directory. The template lives at references/report-templates/ (sibling to references/optimization-report/, NOT inside it). Re-read this file's HTML Generation section rather than concluding no template exists.

Purpose

Create the final structured JSON, markdown summary, and static HTML report that records the asset, input/output paths, profile metrics, operations, validator findings, optimization verdict, agent reasoning, and high-level Stage Optimization Score. The report title presented to readers is USD Performance Tuning Report.

Prerequisites

Baseline and after profile results.
compare-profiles verdict and per-metric comparison.
Ordered operations performed and their outcomes.
Validator findings, including clean checks.
Output path for the optimized stage.
Measurement context for stage/composition metrics (profiling mode, runtime, cache state, sample count, and stage-open method when known).
Optional runtime profiling handoff details when Omniperf or another runtime profiler produced a dashboard/artifact.

Examples

"Generate the optimization report from these profile comparisons and validator results."
"Create the final JSON, markdown summary, and static HTML report for this optimized USD."

Inputs

Collect from prior steps:

Baseline profile (from profile-stage before).
After profile (from profile-stage after).
Compare-profiles verdict and per-metric results.
Operations performed — ordered list with method and outcome.
Validator findings — what was checked, what was found, what was clean.
Output file path — where the optimized stage was saved.
Runtime context — copy the runtime_context object from <output_path>/setup-preflight.json verbatim (canonical location; see skills/omniverse-usd-performance-tuning/references/setup-usd-performance-tuning/references/runtime-context-header.md Where artifacts live). The report must record exactly which Kit application, Scene Optimizer version, and Asset Validator version produced the result so a later reader can reproduce or audit the run.
Measurement context — for the stage/composition measurements used in the score.
Reasoning — one to two concise paragraphs explaining why this specific optimization approach was chosen for this asset, including evidence and tradeoffs.
Runtime profiling handoff — if runtime metrics matter, link or reference the Omniperf dashboard/artifacts separately instead of mixing RAM, VRAM, FPS, frame time, shader, or renderer data into the stage score.

Output Format

Produce three artifacts:

<output_dir>/<asset_name>_optimization_report.json — structured, parseable.
<output_dir>/<asset_name>_optimization_report.md — human-readable summary.
<output_dir>/<asset_name>_optimization_report.html — static, styled, self-contained HTML for stakeholder review.

The markdown and HTML are generated from the JSON — the JSON is the source of truth.

Use the static templates under references/report-templates/ when available. Do not use JavaScript charting libraries or external assets. CSS-only score rings, bars, badges, and color blocks are acceptable.

HTML Generation (mandatory)

Do NOT write the HTML report by hand. Always invoke the committed renderer:

python3 references/report-templates/render_preview.py \
  --fixture <output_dir>/<asset_name>_optimization_report.json \
  --output  <output_dir>/<asset_name>_optimization_report.html

The renderer applies the designed template (optimization-report.html.template) with correct score rings, metric cards, evidence tables, and NVIDIA styling. Hand-written or LLM-generated HTML will not match the design system and is non-conformant — the scored_static_html_report_required guardrail requires use of this renderer.

JSON schema

The report JSON must conform to scripts/optimization-report.schema.json.

Structure:

{
  "asset_name": "SnowdonTowers_SampleHVAC",
  "input_path": "/path/to/original.usd",
  "output_path": "/path/to/optimized.usdc",
  "timestamp": "2026-01-01T00:00:00Z",
  "verdict": "improved",
  "runtime_context": {
    "kit": {
      "application": "USD Composer",
      "version": "110.1.0",
      "path": "D:\\build\\chk\\usd_composer-fat\\110.1.0+main.…\\kit",
      "build": "110.1.0+main.10181.f4b28ef2.gl.windows-x86_64.release"
    },
    "sceneOptimizer": {
      "extension": "omni.scene.optimizer.core",
      "version": "110.0.4"
    },
    "assetValidator": {
      "package": "omniverse-asset-validator",
      "version": "1.x.y",
      "source": "kit-extension"
    }
  },
  "optimization_score": 7.8,
  "score_scope": "stage_optimization",
  "score_label": "strong",
  "reasoning": "The agent prioritized composition cleanup because the baseline profile showed high layer count, expensive stage open, and repeated structure. The chosen operations reduce composition and traversal cost without changing visual intent.\n\nMore aggressive mesh repair or decimation was left out because validator findings indicate bounded-loss decisions that need user approval. Runtime profiling is separated into an Omniperf handoff.",
  "measurement_context": {
    "profile_mode": "quick USD composition profile",
    "runtime": "standalone USD Python",
    "score_scope": "stage/composition metrics only"
  },
  "runtime_profiling": {
    "status": "not_run",
    "recommended_tool": "Omniperf",
    "dashboard_url": null,
    "artifact_path": null,
    "summary": "Runtime profiling was not run for this report.",
    "caveat": "Use Omniperf for RAM, VRAM, FPS, frame time, shader, renderer, and GPU metrics."
  },
  "artifacts": {
    "json": "/path/to/SnowdonTowers_SampleHVAC_optimization_report.json",
    "markdown": "/path/to/SnowdonTowers_SampleHVAC_optimization_report.md",
    "html": "/path/to/SnowdonTowers_SampleHVAC_optimization_report.html"
  },
  "metric_groups": [
    {
      "id": "load_time",
      "display_name": "Composition Load",
      "score": 9.0,
      "status": "measured",
      "weight": 35,
      "summary": "Cold and warm open improved strongly."
    },
    {
      "id": "composition",
      "display_name": "Composition Complexity",
      "score": 8.5,
      "status": "measured",
      "weight": 25,
      "summary": "Layer and reference graph complexity improved."
    }
  ],
  "metrics": [
    {
      "name": "file_size_mb",
      "display_name": "File Size",
      "category": "storage_proxy",
      "unit": "MB",
      "before": 114.2,
      "after": 56.1,
      "change_pct": -50.9,
      "verdict": "improved"
    }
  ],
  "operations": [
    {
      "order": 1,
      "name": "Duplicate discipline deactivation",
      "method": "USD Python API (SetActive(False))",
      "result": "Removed 30 of 37 discipline subtrees at identical world transforms"
    }
  ],
  "validators": [
    {
      "name": "FlatHierarchiesChecker",
      "issues": 16,
      "notes": "Prototypes flat list, 16 repeated discipline children"
    },
    {
      "name": "EmptyLeafChecker",
      "issues": 1525,
      "notes": "Candidates for pruneLeaves"
    }
  ]
}

Markdown template

## USD Performance Tuning Report — {asset_name}

Stage Optimization Score: {optimization_score}/10 ({score_label})
Verdict: {verdict}

### Output File
{output_path}

### Reasoning

{reasoning}

### Runtime Context

| Component | Value |
|---|---|
| Kit application | {runtime_context.kit.application} {runtime_context.kit.version} |
| Kit path | {runtime_context.kit.path} |
| Kit build | {runtime_context.kit.build} |
| Scene Optimizer | {runtime_context.sceneOptimizer.extension} {runtime_context.sceneOptimizer.version} |
| Asset Validator | {runtime_context.assetValidator.package} {runtime_context.assetValidator.version} (via {runtime_context.assetValidator.source}) |

---

### Before / After at a Glance

| Metric | Before | After | Change |
|---|---|---|---|
| {for each metric} |

### Stage Impact Areas

| Area | Score | Status | Notes |
|---|---:|---|---|
| Composition Load | {score}/10 | measured | {summary} |
| Composition Complexity | {score}/10 | measured | {summary} |
| Instancing | {score}/10 | measured | {summary} |
| Validation | {score}/10 | measured | {summary} |

### Runtime Profiling

Runtime metrics such as RAM, VRAM, FPS, frame time, shader cost, and renderer
activity are not part of the Stage Optimization Score. Use Omniperf or an
equivalent runtime profiling dashboard for those measurements.

---

### Operations Performed

| # | Operation | Method | Result |
|---|---|---|---|
| {for each operation} |

---

### Validators Attempted / Result

| Validator | Result |
|---|---|
| {for each validator} |

Rules

⚠️ Schema conformance is mandatory — do not improvise the report shape. Read scripts/optimization-report.schema.json in full before writing the report JSON. Do not synthesize field names or array-item structures from partial reads or memory. Each array (metrics, operations, validators) uses flat records with specific required keys — not nested objects with baseline/after/deltas or any other invented layout. If the schema file exceeds your context window, use the inline example in this reference as the canonical shape and validate your output against the required and items blocks before emitting. The schema rejects extra keys in metric_groups[], metrics[], operations[], validators[], and artifacts; update the schema first if a new report field is genuinely needed. Conformance failures are caught by scored_static_html_report_required in runtime checks.

Always save the JSON report — it is the structured record of what happened.
Always present the markdown to the user in chat.
Always produce the static HTML report when writing report artifacts. In runtime checks, include scored_static_html_report_required in phase_guardrails when planning this final report contract.
Always title the reader-facing report USD Performance Tuning Report.
Always include a dedicated Reasoning section with one to two concise paragraphs explaining why the selected optimizations fit the evidence.
Always compute and present a Stage Optimization Score from stage/composition metrics only. Good inputs include open/traverse/attribute-resolution timing, prim/layer/reference counts, instance/prototype coverage, time-sample counts, extent coverage, duplicate geometry/material findings, and validator deltas.
Compute the top-level score deterministically from scored metric groups: round(sum(group.score * group.weight) / sum(group.weight), 1). Exclude groups with score=null or weight=0. Do not hand-edit the top-level score after computing it.
Derive score_label from the computed score: excellent for >= 9.0, strong for >= 7.5, moderate for >= 5.5, neutral for >= 4.5, mixed for >= 2.5, and regressed below 2.5.
Score each metric group with the same rubric: 10 for very large direct improvement (roughly >= 75%), 8.5 for strong improvement (>= 50%), 7 for clear improvement (>= 25%), 6 for small improvement (>= 10%), 5 for neutral/no meaningful change, 3 for mild regression, 1 for major regression, and 0 for failed or unusable output. Use intermediate values only when the evidence falls between bands, and explain that in reasoning or the group caveat.
Do not include RAM, VRAM, FPS, frame time, shader cost, or renderer activity in the Stage Optimization Score. Put those in runtime_profiling and point to Omniperf or equivalent runtime profiler artifacts when available.
If runtime observations are included for context, label them as external profiling observations under a specific configuration, not intrinsic asset properties.
Use structural metrics (file_size_*, prim_count, mesh_count, layer_count, total_vertices, total_attributes, material counts) as stage/composition evidence. Treat file size carefully: compare total referenced footprint, not root-layer size only.
Generate HTML by invoking references/report-templates/render_preview.py with --fixture pointing to the report JSON and --output for the HTML path. Never write HTML directly — always use the renderer (see HTML Generation above).
Include metrics that didn't change (neutral) — this shows coverage.
Include validators that found 0 issues — this shows what was checked.
If the verdict is "regressed" or "mixed", say so clearly — do not hide regressions.
Include the timestamp so results are traceable to a specific run.

Limitations

Does not profile, validate, or optimize the stage.
Depends on upstream data quality; missing operations or validators should be marked as unavailable.
The JSON report is the source of truth, with markdown generated from it.

Troubleshooting

If the schema validation fails, compare the report against scripts/optimization-report.schema.json.
If the verdict is missing, return to compare-profiles before writing the final report.
If an output path is unknown, stop and capture it rather than leaving the deliverable ambiguous.

Requirements and compatibility

Written for: OpenAI Codex

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Needs Python

The author says: Orchestrator skill. Downstream phases may require Kit, Scene Optimizer, Asset Validator, USD Python, writable output paths, and omniverse:// authentication selected by setup-usd-performance-tuning.

Security

What our scanner noticed· Medium risk

Large base64-encoded blobs
Instructions that use the network
Scripts that your agent may run

This skill includes scripts that your agent may run. Read them before you use it.

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by

NVIDIA Omniverse

Publisher

OpenAI (Official)

Distributed by

OpenAI

Source

openai/plugins · plugins/nvidia/skills/omniverse-usd-performance-tuning · commit d420d13

Licence

Apache License 2.0 (redistribution allowed)

LICENSE (written by AutoGineer)
ATTRIBUTION.md (written by AutoGineer)

Synced

2026-11-08

AutoGineer is not affiliated with or endorsed by NVIDIA Omniverse or OpenAI.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download omniverse-usd-performance-tuning.zip

From the author

AutoGineer summary

What it does

When to use it

Advantages

Usage examples

Install

Documentation by NVIDIA Omniverse

Omniverse USD Performance Tuning

When to Use

Instructions

Output Format

Entry skill rule

Runtime context — session-start gate (mandatory)

Runtime artifact token budget

Plan-time vs execution-time approval

Output expectation

Purpose

Prerequisites

Examples

Triage order

Optimization ordering

Rules

Limitations

Troubleshooting

References

Required execution flow

Files

references/optimization-report/README.md

Optimization Report

When to Use

Instructions

Pre-flight Checklist

Anti-pattern: hand-written HTML

Purpose

Prerequisites

Examples

Inputs

Output Format

HTML Generation (mandatory)

JSON schema

Markdown template

Rules

Limitations

Troubleshooting

Requirements and compatibility

Security

Attribution and licence

Related skills

Diagnosing Bugs

Systematic Debugging

Performance Optimization

Debugging And Error Recovery

Animation Performance

Parallel Debugging

Download ZIP