Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Unity Profiler skill for performance analysis, frame debugging, memory profiling, and optimization workflows.
.claude/skills/a5c-ai-unity-profiler/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-02 | ✓→✗ | ▼ Worse | 15% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 6% | 0% |
Performance profiling and optimization using Unity Profiler tools.
This skill provides capabilities for analyzing and optimizing Unity game performance using the Profiler, Frame Debugger, and Memory Profiler.
csharpusing Unity.Profiling; public class OptimizedSystem : MonoBehaviour { static readonly ProfilerMarker s_UpdateMarker = new ProfilerMarker("MySystem.Update"); void Update() { using (s_UpdateMarker.Auto()) { // Code to profile ProcessEntities(); } } }
csharp[Test, Performance] public void TestSpawnPerformance() { Measure.Method(() => { for (int i = 0; i < 1000; i++) { SpawnEnemy(); } }) .WarmupCount(3) .MeasurementCount(10) .Run(); }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 28,498 | 22,901 | -20% | 1 | 1 | 0% | 3,776 | 3,389 | -10% | 0 | 0 | — |
case-02 | pass→fail | 13,842 | 12,672 | -8% | 1 | 1 | 0% | 2,247 | 2,586 | +15% | 0 | 0 | — |
case-03 | pass→pass | 5,079 | 2,607 | -49% | 1 | 1 | 0% | 646 | 817 | +26% | 0 | 0 | — |
case-04 | pass→pass | 9,862 | 8,218 | -17% | 1 | 1 | 0% | 1,672 | 1,637 | -2% | 0 | 0 | — |
case-05 | fail→pass | 7,145 | 5,663 | -21% | 1 | 1 | 0% | 1,240 | 1,290 | +4% | 0 | 0 | — |
case-06 | fail→pass | 3,411 | 1,798 | -47% | 1 | 1 | 0% | 477 | 629 | +32% | 0 | 0 | — |
case-07 | pass→pass | 7,216 | 3,918 | -46% | 1 | 1 | 0% | 1,117 | 963 | -14% | 0 | 0 | — |
case-08 | pass→pass | 19,264 | 9,521 | -51% | 1 | 1 | 0% | 2,626 | 1,795 | -32% | 0 | 0 | — |
case-09 | pass→pass | 19,833 | 15,301 | -23% | 1 | 1 | 0% | 2,654 | 2,863 | +8% | 0 | 0 | — |
case-10 | pass→pass | 10,452 | 7,928 | -24% | 1 | 1 | 0% | 1,686 | 1,761 | +4% | 0 | 0 | — |
case-11 | pass→pass | 17,221 | 8,121 | -53% | 1 | 1 | 0% | 2,225 | 1,670 | -25% | 0 | 0 | — |
case-12 | pass→pass | 17,682 | 13,076 | -26% | 1 | 1 | 0% | 2,320 | 2,435 | +5% | 0 | 0 | — |
case-13 | pass→pass | 17,780 | 13,310 | -25% | 1 | 1 | 0% | 2,473 | 2,633 | +6% | 0 | 0 | — |
case-14 | pass→pass | 18,378 | 20,461 | +11% | 1 | 1 | 0% | 2,951 | 3,338 | +13% | 0 | 0 | — |
case-15 | pass→pass | 13,856 | 14,244 | +3% | 1 | 1 | 0% | 2,130 | 2,710 | +27% | 0 | 0 | — |
case-16 | fail→fail | 5,545 | 6,236 | +12% | 1 | 1 | 0% | 896 | 1,310 | +46% | 0 | 0 | — |
case-17 | pass→pass | 15,280 | 9,969 | -35% | 1 | 1 | 0% | 2,268 | 1,926 | -15% | 0 | 0 | — |
case-18 | pass→pass | 4,771 | 3,742 | -22% | 1 | 1 | 0% | 727 | 1,038 | +43% | 0 | 0 | — |
case-19 | pass→pass | 5,360 | 4,212 | -21% | 1 | 1 | 0% | 893 | 988 | +11% | 0 | 0 | — |
case-20 | pass→pass | 19,182 | 20,351 | +6% | 1 | 1 | 0% | 3,499 | 3,526 | +1% | 0 | 0 | — |
case-21 | pass→pass | 16,076 | 11,807 | -27% | 1 | 1 | 0% | 2,350 | 2,505 | +7% | 0 | 0 | — |
case-22 | pass→fail | 21,301 | 15,535 | -27% | 1 | 1 | 0% | 2,801 | 2,964 | +6% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.