Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Unreal Engine C++ programming skill for UCLASS macros, reflection system, garbage collection, and engine integration.
.claude/skills/a5c-ai-unreal-cpp/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✓→✓ | = Same ✓ | 1% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 15% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 5% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 29% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 28% | 0% |
C++ programming for Unreal Engine development.
This skill provides capabilities for implementing Unreal Engine gameplay and systems using C++, including the reflection system, macros, and engine integration.
cppUCLASS() class MYGAME_API AMyCharacter : public ACharacter { GENERATED_BODY() public: AMyCharacter(); UPROPERTY(EditAnywhere, BlueprintReadWrite, Category = "Stats") float Health = 100.f; UFUNCTION(BlueprintCallable, Category = "Combat") void TakeDamage(float Damage); protected: virtual void BeginPlay() override; virtual void Tick(float DeltaTime) override; };
cppUCLASS(ClassGroup=(Custom), meta=(BlueprintSpawnableComponent)) class MYGAME_API UHealthComponent : public UActorComponent { GENERATED_BODY() public: UPROPERTY(BlueprintAssignable) FOnHealthChanged OnHealthChanged; UFUNCTION(BlueprintCallable) void ModifyHealth(float Delta); private: UPROPERTY(EditDefaultsOnly) float MaxHealth = 100.f; float CurrentHealth; };
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 9,945 | 34,152 | +243% | 1 | 1 | 0% | 1,479 | 1,497 | +1% | 0 | 0 | — |
case-02 | fail→fail | 16,269 | 11,432 | -30% | 1 | 1 | 0% | 2,507 | 2,532 | +1% | 0 | 0 | — |
case-03 | pass→pass | 12,484 | 12,413 | -1% | 1 | 1 | 0% | 2,492 | 2,872 | +15% | 0 | 0 | — |
case-04 | pass→pass | 10,183 | 5,445 | -47% | 1 | 1 | 0% | 1,357 | 1,424 | +5% | 0 | 0 | — |
case-05 | pass→pass | 11,188 | 11,308 | +1% | 1 | 1 | 0% | 2,057 | 2,657 | +29% | 0 | 0 | — |
case-06 | pass→pass | 9,952 | 10,300 | +3% | 1 | 1 | 0% | 1,535 | 1,967 | +28% | 0 | 0 | — |
case-07 | pass→pass | 7,850 | 6,189 | -21% | 1 | 1 | 0% | 1,278 | 1,449 | +13% | 0 | 0 | — |
case-08 | pass→pass | 12,399 | 10,405 | -16% | 1 | 1 | 0% | 2,167 | 2,327 | +7% | 0 | 0 | — |
case-09 | pass→pass | 9,384 | 9,730 | +4% | 1 | 1 | 0% | 1,596 | 1,770 | +11% | 0 | 0 | — |
case-10 | pass→pass | 7,195 | 5,019 | -30% | 1 | 1 | 0% | 1,211 | 1,268 | +5% | 0 | 0 | — |
case-11 | pass→pass | 12,219 | 8,791 | -28% | 1 | 1 | 0% | 1,838 | 1,973 | +7% | 0 | 0 | — |
case-12 | pass→pass | 7,800 | 5,301 | -32% | 1 | 1 | 0% | 1,354 | 1,367 | +1% | 0 | 0 | — |
case-13 | pass→pass | 7,327 | 6,816 | -7% | 1 | 1 | 0% | 1,219 | 1,302 | +7% | 0 | 0 | — |
case-14 | pass→pass | 13,807 | 7,934 | -43% | 1 | 1 | 0% | 1,884 | 1,660 | -12% | 0 | 0 | — |
case-15 | pass→pass | 14,571 | 12,737 | -13% | 1 | 1 | 0% | 2,079 | 2,343 | +13% | 0 | 0 | — |
case-16 | pass→pass | 5,089 | 3,318 | -35% | 1 | 1 | 0% | 963 | 1,090 | +13% | 0 | 0 | — |
case-17 | pass→pass | 19,680 | 12,788 | -35% | 1 | 1 | 0% | 1,893 | 2,179 | +15% | 0 | 0 | — |
case-18 | pass→pass | 19,568 | 15,043 | -23% | 1 | 1 | 0% | 2,817 | 3,366 | +19% | 0 | 0 | — |
case-19 | pass→pass | 7,948 | 6,786 | -15% | 1 | 1 | 0% | 1,414 | 1,819 | +29% | 0 | 0 | — |
case-20 | pass→pass | 8,463 | 5,992 | -29% | 1 | 1 | 0% | 1,111 | 1,298 | +17% | 0 | 0 | — |
case-21 | pass→pass | 10,853 | 7,308 | -33% | 1 | 1 | 0% | 1,550 | 1,605 | +4% | 0 | 0 | — |
case-22 | pass→pass | 10,125 | 7,091 | -30% | 1 | 1 | 0% | 1,467 | 1,720 | +17% | 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 -100 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Other measured skills in the registry, with their headline benchmark lift.