Install any skill in seconds. Free to start, no credit card required.
Get Started Free →High Definition Render Pipeline configuration for Unity, including ray tracing, volumetric effects, and high-fidelity graphics setup.
.claude/skills/a5c-ai-unity-hdrp/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-24 | ✓→✗ | ▼ Worse | 16% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 31% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 26% | 0% |
High Definition Render Pipeline configuration for high-fidelity graphics in Unity.
This skill provides capabilities for configuring and extending Unity's High Definition Render Pipeline, enabling ray tracing, volumetric effects, and AAA-quality graphics.
csharp// Configure HDRP Lit material var material = new Material(Shader.Find("HDRP/Lit")); material.SetFloat("_Metallic", 0.8f); material.SetFloat("_Smoothness", 0.9f); material.EnableKeyword("_NORMALMAP");
1. Create Volume (Global or Local)
2. Add Volume Profile
3. Add overrides (Fog, Exposure, etc.)
4. Configure priority and blend distance| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 10,159 | 28,442 | +180% | 1 | 1 | 0% | 1,827 | 2,400 | +31% | 0 | 0 | — |
case-02 | pass→pass | 10,995 | 11,199 | +2% | 1 | 1 | 0% | 1,905 | 2,399 | +26% | 0 | 0 | — |
case-03 | pass→pass | 9,980 | 11,839 | +19% | 1 | 1 | 0% | 1,739 | 2,309 | +33% | 0 | 0 | — |
case-04 | fail→fail | 14,681 | 10,984 | -25% | 1 | 1 | 0% | 2,177 | 2,379 | +9% | 0 | 0 | — |
case-05 | pass→pass | 16,848 | 15,595 | -7% | 1 | 1 | 0% | 2,763 | 3,274 | +18% | 0 | 0 | — |
case-06 | pass→pass | 15,183 | 16,810 | +11% | 1 | 1 | 0% | 2,354 | 2,754 | +17% | 0 | 0 | — |
case-07 | pass→pass | 20,945 | 23,742 | +13% | 1 | 1 | 0% | 3,637 | 4,316 | +19% | 0 | 0 | — |
case-08 | pass→pass | 20,635 | 13,470 | -35% | 1 | 1 | 0% | 2,957 | 2,746 | -7% | 0 | 0 | — |
case-09 | pass→pass | 18,141 | 16,776 | -8% | 1 | 1 | 0% | 2,853 | 3,377 | +18% | 0 | 0 | — |
case-10 | fail→pass | 10,116 | 6,873 | -32% | 1 | 1 | 0% | 1,399 | 1,562 | +12% | 0 | 0 | — |
case-11 | pass→pass | 15,633 | 18,521 | +18% | 1 | 1 | 0% | 2,486 | 3,505 | +41% | 0 | 0 | — |
case-12 | pass→pass | 16,239 | 13,711 | -16% | 1 | 1 | 0% | 2,470 | 2,894 | +17% | 0 | 0 | — |
case-13 | pass→pass | 18,262 | 18,870 | +3% | 1 | 1 | 0% | 2,874 | 3,183 | +11% | 0 | 0 | — |
case-14 | pass→pass | 29,312 | 21,942 | -25% | 1 | 1 | 0% | 2,720 | 3,579 | +32% | 0 | 0 | — |
case-15 | pass→pass | 7,497 | 6,743 | -10% | 1 | 1 | 0% | 911 | 1,607 | +76% | 0 | 0 | — |
case-16 | pass→pass | 16,212 | 15,136 | -7% | 1 | 1 | 0% | 2,280 | 3,211 | +41% | 0 | 0 | — |
case-17 | fail→fail | 18,995 | 16,160 | -15% | 1 | 1 | 0% | 2,610 | 2,960 | +13% | 0 | 0 | — |
case-18 | pass→pass | 6,384 | 5,023 | -21% | 1 | 1 | 0% | 773 | 1,118 | +45% | 0 | 0 | — |
case-19 | pass→pass | 9,018 | 8,163 | -9% | 1 | 1 | 0% | 1,396 | 1,835 | +31% | 0 | 0 | — |
case-20 | fail→pass | 13,531 | 2,479 | -82% | 1 | 1 | 0% | 2,276 | 880 | -61% | 0 | 0 | — |
case-21 | pass→pass | 14,151 | 13,677 | -3% | 1 | 1 | 0% | 2,204 | 2,451 | +11% | 0 | 0 | — |
case-22 | pass→pass | 16,795 | 18,405 | +10% | 1 | 1 | 0% | 3,173 | 4,000 | +26% | 0 | 0 | — |
case-23 | pass→pass | 9,190 | 10,096 | +10% | 1 | 1 | 0% | 1,578 | 1,983 | +26% | 0 | 0 | — |
case-24 | pass→fail | 9,741 | 9,939 | +2% | 1 | 1 | 0% | 1,644 | 1,900 | +16% | 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. 24 cases were attempted. The headline lift of +4 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is 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.