Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Godot C# programming skill for .NET integration, scripting patterns, and performance optimization.
.claude/skills/a5c-ai-godot-csharp/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | 120% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 11% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 60% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 26% | 0% |
| case-04 | ✓→✓ | = Same ✓ | -1% | 0% |
C# programming for Godot Engine development.
This skill provides capabilities for implementing game logic using C# in Godot, leveraging .NET integration.
csharpusing Godot; public partial class Player : CharacterBody2D { [Export] public float Speed { get; set; } = 200f; [Signal] public delegate void HealthChangedEventHandler(int newHealth); private int _health = 100; public override void _Ready() { // Initialize } public override void _PhysicsProcess(double delta) { var velocity = Vector2.Zero; velocity.X = Input.GetAxis("move_left", "move_right"); velocity.Y = Input.GetAxis("move_up", "move_down"); Velocity = velocity.Normalized() * Speed; MoveAndSlide(); } }
csharpbutton.Pressed += OnButtonPressed; // or button.Connect("pressed", Callable.From(OnButtonPressed));
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 10,052 | 17,667 | +76% | 1 | 1 | 0% | 1,617 | 1,787 | +11% | 0 | 0 | — |
case-02 | pass→pass | 4,566 | 4,576 | +0% | 1 | 1 | 0% | 796 | 1,274 | +60% | 0 | 0 | — |
case-03 | pass→pass | 3,801 | 3,036 | -20% | 1 | 1 | 0% | 784 | 991 | +26% | 0 | 0 | — |
case-04 | pass→pass | 7,345 | 5,539 | -25% | 1 | 1 | 0% | 1,350 | 1,333 | -1% | 0 | 0 | — |
case-05 | pass→pass | 16,704 | 13,093 | -22% | 1 | 1 | 0% | 2,661 | 2,987 | +12% | 0 | 0 | — |
case-06 | pass→pass | 9,936 | 7,709 | -22% | 1 | 1 | 0% | 2,163 | 1,999 | -8% | 0 | 0 | — |
case-07 | pass→pass | 2,384 | 2,735 | +15% | 1 | 1 | 0% | 436 | 861 | +97% | 0 | 0 | — |
case-08 | pass→pass | 11,370 | 11,850 | +4% | 1 | 1 | 0% | 1,976 | 2,465 | +25% | 0 | 0 | — |
case-09 | pass→pass | 11,182 | 8,245 | -26% | 1 | 1 | 0% | 2,093 | 2,079 | -1% | 0 | 0 | — |
case-10 | pass→pass | 6,668 | 3,164 | -53% | 1 | 1 | 0% | 1,002 | 977 | -2% | 0 | 0 | — |
case-11 | pass→pass | 9,577 | 9,216 | -4% | 1 | 1 | 0% | 1,636 | 1,803 | +10% | 0 | 0 | — |
case-12 | pass→pass | 19,436 | 21,261 | +9% | 1 | 1 | 0% | 3,395 | 3,915 | +15% | 0 | 0 | — |
case-13 | pass→pass | 17,018 | 15,964 | -6% | 1 | 1 | 0% | 3,178 | 3,418 | +8% | 0 | 0 | — |
case-14 | pass→pass | 3,389 | 4,293 | +27% | 1 | 1 | 0% | 545 | 1,192 | +119% | 0 | 0 | — |
case-15 | pass→pass | 9,787 | 9,931 | +1% | 1 | 1 | 0% | 1,980 | 2,073 | +5% | 0 | 0 | — |
case-16 | pass→pass | 7,160 | 3,794 | -47% | 1 | 1 | 0% | 1,251 | 1,148 | -8% | 0 | 0 | — |
case-17 | pass→pass | 3,530 | 2,549 | -28% | 1 | 1 | 0% | 559 | 919 | +64% | 0 | 0 | — |
case-18 | pass→pass | 11,271 | 10,445 | -7% | 1 | 1 | 0% | 1,800 | 2,299 | +28% | 0 | 0 | — |
case-19 | pass→pass | 7,369 | 5,572 | -24% | 1 | 1 | 0% | 1,114 | 1,519 | +36% | 0 | 0 | — |
case-20 | fail→pass | 3,122 | 3,430 | +10% | 1 | 1 | 0% | 495 | 1,088 | +120% | 0 | 0 | — |
case-21 | pass→pass | 15,059 | 10,825 | -28% | 1 | 1 | 0% | 2,432 | 2,826 | +16% | 0 | 0 | — |
case-22 | pass→pass | 2,965 | 4,094 | +38% | 1 | 1 | 0% | 538 | 1,231 | +129% | 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.