Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Unity New Input System configuration skill for action maps, device bindings, control schemes, and cross-platform input handling.
.claude/skills/a5c-ai-unity-input-system/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-02 | ✓→✓ | = Same ✓ | -4% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 4% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 6% | 0% |
New Input System configuration and implementation for Unity.
This skill provides capabilities for implementing Unity's New Input System, including action maps, device support, and cross-platform input handling.
csharp// Generated PlayerInput class usage public class PlayerController : MonoBehaviour { private PlayerInputActions inputActions; void Awake() { inputActions = new PlayerInputActions(); } void OnEnable() { inputActions.Gameplay.Enable(); inputActions.Gameplay.Jump.performed += OnJump; inputActions.Gameplay.Move.performed += OnMove; } void OnDisable() { inputActions.Gameplay.Disable(); } void OnJump(InputAction.CallbackContext ctx) { // Handle jump } void OnMove(InputAction.CallbackContext ctx) { Vector2 movement = ctx.ReadValue<Vector2>(); } }
csharppublic void StartRebinding(InputAction action) { action.PerformInteractiveRebinding() .WithControlsExcluding("Mouse") .OnComplete(operation => { operation.Dispose(); SaveBindings(); }) .Start(); }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 13,443 | 8,936 | -34% | 1 | 1 | 0% | 2,144 | 2,268 | +6% | 0 | 0 | — |
case-02 | pass→pass | 12,222 | 8,748 | -28% | 1 | 1 | 0% | 2,217 | 2,126 | -4% | 0 | 0 | — |
case-03 | pass→pass | 10,067 | 7,453 | -26% | 1 | 1 | 0% | 1,570 | 1,637 | +4% | 0 | 0 | — |
case-04 | pass→pass | 22,424 | 15,726 | -30% | 1 | 1 | 0% | 2,766 | 2,919 | +6% | 0 | 0 | — |
case-05 | pass→pass | 10,758 | 6,282 | -42% | 1 | 1 | 0% | 1,605 | 1,603 | -0% | 0 | 0 | — |
case-06 | pass→pass | 20,072 | 12,329 | -39% | 1 | 1 | 0% | 2,943 | 2,613 | -11% | 0 | 0 | — |
case-07 | pass→pass | 4,309 | 4,683 | +9% | 1 | 1 | 0% | 679 | 1,143 | +68% | 0 | 0 | — |
case-08 | pass→pass | 7,606 | 7,817 | +3% | 1 | 1 | 0% | 1,347 | 1,548 | +15% | 0 | 0 | — |
case-09 | pass→pass | 18,613 | 14,181 | -24% | 1 | 1 | 0% | 2,941 | 3,128 | +6% | 0 | 0 | — |
case-10 | pass→pass | 5,436 | 5,681 | +5% | 1 | 1 | 0% | 973 | 1,290 | +33% | 0 | 0 | — |
case-11 | pass→pass | 18,748 | 15,501 | -17% | 1 | 1 | 0% | 2,872 | 2,817 | -2% | 0 | 0 | — |
case-12 | pass→pass | 12,751 | 10,443 | -18% | 1 | 1 | 0% | 2,065 | 2,139 | +4% | 0 | 0 | — |
case-13 | pass→pass | 9,792 | 9,041 | -8% | 1 | 1 | 0% | 1,773 | 2,143 | +21% | 0 | 0 | — |
case-14 | fail→pass | 10,700 | 9,026 | -16% | 1 | 1 | 0% | 1,377 | 1,703 | +24% | 0 | 0 | — |
case-15 | pass→pass | 12,802 | 13,239 | +3% | 1 | 1 | 0% | 1,821 | 2,173 | +19% | 0 | 0 | — |
case-16 | pass→pass | 11,189 | 8,246 | -26% | 1 | 1 | 0% | 1,529 | 1,955 | +28% | 0 | 0 | — |
case-17 | fail→pass | 6,356 | 3,832 | -40% | 1 | 1 | 0% | 899 | 1,017 | +13% | 0 | 0 | — |
case-18 | pass→pass | 5,924 | 6,642 | +12% | 1 | 1 | 0% | 914 | 1,375 | +50% | 0 | 0 | — |
case-19 | pass→pass | 9,675 | 4,373 | -55% | 1 | 1 | 0% | 1,464 | 1,325 | -9% | 0 | 0 | — |
case-20 | pass→pass | 8,771 | 11,523 | +31% | 1 | 1 | 0% | 1,693 | 2,154 | +27% | 0 | 0 | — |
case-21 | pass→pass | 13,776 | 12,206 | -11% | 1 | 1 | 0% | 2,149 | 2,820 | +31% | 0 | 0 | — |
case-22 | pass→pass | 12,854 | 8,490 | -34% | 1 | 1 | 0% | 1,869 | 2,134 | +14% | 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 +9 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.