Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Unreal Engine MetaSounds skill for procedural audio, real-time synthesis, and advanced audio graphs.
.claude/skills/a5c-ai-unreal-metasounds/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -8% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 13% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 12% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | -13% | 0% |
MetaSounds procedural audio system for Unreal Engine.
This skill provides capabilities for implementing procedural audio using Unreal's MetaSounds system.
1. Create MetaSound Source
2. Add input parameters
3. Build audio graph
4. Configure output
5. Use in Sound Cue or directlycppUAudioComponent* AudioComp = CreateAudioComponent(); AudioComp->SetFloatParameter(FName("Intensity"), 0.8f); AudioComp->SetTriggerParameter(FName("OnHit"));
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 28,817 | 27,074 | -6% | 1 | 1 | 0% | 4,606 | 3,594 | -22% | 0 | 0 | — |
case-02 | pass→pass | 26,644 | 17,595 | -34% | 1 | 1 | 0% | 3,881 | 3,584 | -8% | 0 | 0 | — |
case-03 | pass→pass | 13,860 | 16,482 | +19% | 1 | 1 | 0% | 2,352 | 2,647 | +13% | 0 | 0 | — |
case-04 | pass→pass | 18,517 | 17,720 | -4% | 1 | 1 | 0% | 2,647 | 2,972 | +12% | 0 | 0 | — |
case-05 | pass→pass | 14,078 | 8,660 | -38% | 1 | 1 | 0% | 2,059 | 1,797 | -13% | 0 | 0 | — |
case-06 | pass→pass | 14,974 | 8,792 | -41% | 1 | 1 | 0% | 1,992 | 1,528 | -23% | 0 | 0 | — |
case-07 | pass→pass | 8,870 | 6,817 | -23% | 1 | 1 | 0% | 1,197 | 1,211 | +1% | 0 | 0 | — |
case-08 | fail→pass | 9,239 | 6,822 | -26% | 1 | 1 | 0% | 1,074 | 1,243 | +16% | 0 | 0 | — |
case-09 | fail→fail | 14,396 | 21,393 | +49% | 1 | 1 | 0% | 2,789 | 3,570 | +28% | 0 | 0 | — |
case-10 | pass→pass | 4,860 | 1,398 | -71% | 1 | 1 | 0% | 624 | 462 | -26% | 0 | 0 | — |
case-11 | pass→pass | 3,393 | 3,277 | -3% | 1 | 1 | 0% | 437 | 733 | +68% | 0 | 0 | — |
case-12 | pass→pass | 15,258 | 12,741 | -16% | 1 | 1 | 0% | 2,316 | 1,961 | -15% | 0 | 0 | — |
case-13 | pass→pass | 20,054 | 14,851 | -26% | 1 | 1 | 0% | 2,628 | 2,458 | -6% | 0 | 0 | — |
case-14 | pass→pass | 14,132 | 10,708 | -24% | 1 | 1 | 0% | 1,834 | 1,978 | +8% | 0 | 0 | — |
case-15 | pass→pass | 16,148 | 4,774 | -70% | 1 | 1 | 0% | 2,374 | 923 | -61% | 0 | 0 | — |
case-16 | pass→pass | 16,941 | 16,372 | -3% | 1 | 1 | 0% | 2,717 | 2,727 | +0% | 0 | 0 | — |
case-17 | pass→pass | 17,826 | 19,305 | +8% | 1 | 1 | 0% | 2,609 | 2,768 | +6% | 0 | 0 | — |
case-18 | pass→pass | 22,542 | 21,581 | -4% | 1 | 1 | 0% | 3,024 | 3,731 | +23% | 0 | 0 | — |
case-19 | pass→pass | 16,724 | 11,081 | -34% | 1 | 1 | 0% | 2,307 | 2,186 | -5% | 0 | 0 | — |
case-20 | pass→pass | 7,712 | 11,830 | +53% | 1 | 1 | 0% | 1,271 | 1,934 | +52% | 0 | 0 | — |
case-21 | pass→pass | 18,292 | 19,042 | +4% | 1 | 1 | 0% | 2,293 | 2,890 | +26% | 0 | 0 | — |
case-22 | pass→pass | 16,607 | 14,352 | -14% | 1 | 1 | 0% | 2,544 | 2,350 | -8% | 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.
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.