{ "version": "0.1.0", "skill": "omniverse-usd-performance-tuning", "cases": [ { "id": "usd-performance-broad-optimization-flow", "question": "The main factory USD stage takes minutes to open in USD Composer and likely has repeated assemblies. Please optimize it so it loads faster and produces a clear report.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "The agent should select the USD performance tuning workflow, start with runtime/auth bring-up, then plan baseline profiling, structure assessment, validation routing, restructure decision handling, Scene Optimizer operation planning when available, after-profile comparison, and the required optimization report.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Classifies the request as a broad USD performance optimization request rather than a direct single-operation command.", "Starts with the mandatory runtime context gate before profiling, validation, or mutation.", "Plans baseline profiling and usd-structure-assessment before any mutating optimization step.", "Includes validation routing before mutation and an optimization-report deliverable after verification." ] }, { "id": "usd-performance-runtime-choice-gate", "question": "Optimize /tmp/factory.usd. I have not said whether to use Kit or standalone libraries, and there is no existing runtime probe result.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "The agent should not silently pick Kit or standalone. It should route to setup-usd-performance-tuning and ask for the runtime choice before opening, profiling, validating, or optimizing the stage.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Recognizes that runtime choice is unresolved.", "Routes to setup-usd-performance-tuning for the runtime chooser.", "Asks for an explicit Kit or standalone runtime path before doing work.", "Does not open, profile, validate, or mutate the stage before the runtime gate is resolved." ] }, { "id": "usd-performance-destructive-op-approval", "question": "Flatten /tmp/factory.usd into one layer and decimate the high-poly meshes so it runs faster in our Kit-based viewer.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "The agent should treat flattening and decimation as explicitly requested destructive operations, ask for approval before mutation, and preserve the decimation guardrails from the operation reference.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Sets the planning decision to require approval before destructive mutation.", "Names flattening and decimation in the approval prompt.", "Plans structural assessment and validation before mutation if approval is granted.", "Uses one upfront decimation prompt and preserves pinBoundaries plus floating-point stop-condition values." ] }, { "id": "usd-performance-expensive-validation-approval", "question": "Optimize /tmp/factory.usd. The structure assessment suggests occlusion checks and cross-component duplicate checks may be useful before optimization.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "The agent should plan targeted validation and ask before expensive cross-component validators such as occlusion, coincident-geometry, or duplicate-analysis checks run on a large or unknown-size asset.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Routes validation through usd-validation-runner and its validation scoping guidance.", "Treats expensive cross-component checks as opt-in.", "Asks the user before running slow Tier 3-style validation work.", "Keeps the final workflow anchored on an optimization-report deliverable." ] }, { "id": "usd-performance-viewer-build-distractor-negative", "question": "Build a browser-based RTX USD viewer with camera controls, object picking, a stage tree, and render settings.", "expected_skill": null, "expected_script": null, "ground_truth": "This is an application/viewer construction request, not a USD performance diagnosis or optimization workflow.", "expected_behavior": [ "Does not select omniverse-usd-performance-tuning.", "Does not run profiling, validation, Scene Optimizer operations, or optimization-report steps.", "Routes to a viewer or application-building skill if one is available." ] }, { "id": "usd-performance-structural-only-report", "question": "Optimize /tmp/factory.usd, but Scene Optimizer is not installed in my runtime and I don't want to install anything right now. Still give me a report.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "With Scene Optimizer unavailable and install declined, the workflow runs in structural-only mode: structure assessment plus pre-mutation USD validation, no mesh operations. The final report keeps verdict within its enum (neutral if nothing changed), sets workflow_mode to structural_only, and records the Scene Optimizer blocker in notes. It must not invent a structural-only verdict value.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Recognizes Scene Optimizer is unavailable and install was declined, and continues in structural-only mode rather than halting or fabricating results.", "Skips the Scene Optimizer mesh-operation phases.", "Produces a report whose verdict stays within improved|neutral|regressed|mixed and sets workflow_mode to structural_only.", "Records the Scene Optimizer blocker and the next profile capture needed in the report notes." ] }, { "id": "usd-performance-report-schema-conformance", "question": "Finish the optimization run on /tmp/factory.usd and write the final report.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "The final optimization report must conform to optimization-report.schema.json. The agent should validate the report JSON with python3 scripts/validate_report.py before treating it as final, generate HTML via the renderer with --fixture/--output, and must not emit out-of-enum verdicts or invented top-level fields.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md and the optimization-report reference.", "Generates the report JSON against optimization-report.schema.json with verdict in improved|neutral|regressed|mixed.", "Validates the finished report with python3 scripts/validate_report.py before finishing.", "Generates the HTML via render_preview.py with --fixture and --output, not by hand and not argless.", "Does not invent verdict values such as structural-only or no-op-needed." ] }, { "id": "usd-performance-no-overwrite-source", "question": "Run mesh cleanup and decimation on /data/asset.usd to make it lighter.", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "Destructive Scene Optimizer operations must write a new optimized output path and must not overwrite the source asset in place. The agent should also ask for approval before the explicitly named destructive operations.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Asks for approval before the explicitly named destructive operations (mesh cleanup, decimation).", "Saves optimized output to a new path and does not overwrite the original source asset.", "Validates before and after the operations and records the output path in the optimization report." ] }, { "id": "usd-performance-zero-work-operation", "question": "I ran the optimization and the operation report says it completed - did anything actually change?", "expected_skill": "omniverse-usd-performance-tuning", "expected_script": null, "ground_truth": "A Scene Optimizer operation can return success while changing nothing (zero prims affected). The agent should detect a successful-but-no-op result by comparing before/after metrics, report it honestly as neutral, and not claim an improvement the metrics do not support.", "expected_behavior": [ "Reads the omniverse-usd-performance-tuning SKILL.md.", "Compares before/after metrics rather than trusting the operation's success status alone.", "Recognizes a success-but-zero-work result (no prims or meshes changed) and reports it as neutral.", "Does not present an unchanged stage as an improvement in the optimization report." ] } ] }