Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Unreal Engine Lumen skill for global illumination, reflections, and dynamic lighting.
.claude/skills/a5c-ai-unreal-lumen/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 34% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 3% | 0% |
Lumen global illumination and reflections for Unreal Engine.
This skill provides capabilities for implementing Lumen dynamic global illumination and reflections in Unreal Engine.
Lumen Global Illumination:
- Enable in Project Settings
- Configure final gather quality
- Set trace distance
- Configure scene detail1. Add Post Process Volume
2. Enable Infinite Extent
3. Configure Lumen settings
4. Adjust quality per area| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 27,245 | 33,432 | +23% | 1 | 1 | 0% | 3,848 | 4,868 | +27% | 0 | 0 | — |
case-02 | pass→pass | 12,110 | 10,455 | -14% | 1 | 1 | 0% | 2,026 | 2,080 | +3% | 0 | 0 | — |
case-03 | pass→fail | 16,799 | 21,609 | +29% | 1 | 1 | 0% | 2,623 | 3,510 | +34% | 0 | 0 | — |
case-04 | fail→fail | 19,735 | 16,133 | -18% | 1 | 1 | 0% | 2,954 | 3,271 | +11% | 0 | 0 | — |
case-05 | fail→fail | 23,134 | 19,677 | -15% | 1 | 1 | 0% | 3,039 | 2,996 | -1% | 0 | 0 | — |
case-06 | fail→pass | 13,711 | 10,175 | -26% | 1 | 1 | 0% | 2,220 | 2,046 | -8% | 0 | 0 | — |
case-07 | pass→pass | 16,798 | 15,186 | -10% | 1 | 1 | 0% | 2,583 | 2,729 | +6% | 0 | 0 | — |
case-08 | pass→pass | 18,631 | 20,063 | +8% | 1 | 1 | 0% | 3,144 | 3,385 | +8% | 0 | 0 | — |
case-09 | fail→fail | 12,354 | 8,139 | -34% | 1 | 1 | 0% | 2,133 | 1,795 | -16% | 0 | 0 | — |
case-10 | pass→pass | 15,586 | 12,786 | -18% | 1 | 1 | 0% | 2,459 | 2,304 | -6% | 0 | 0 | — |
case-11 | pass→pass | 8,929 | 7,186 | -20% | 1 | 1 | 0% | 1,486 | 1,556 | +5% | 0 | 0 | — |
case-12 | pass→pass | 15,997 | 11,507 | -28% | 1 | 1 | 0% | 2,893 | 2,303 | -20% | 0 | 0 | — |
case-13 | pass→pass | 15,312 | 13,642 | -11% | 1 | 1 | 0% | 2,442 | 2,454 | +0% | 0 | 0 | — |
case-14 | fail→fail | 11,320 | 11,144 | -2% | 1 | 1 | 0% | 1,810 | 2,235 | +23% | 0 | 0 | — |
case-15 | fail→fail | 19,233 | 17,959 | -7% | 1 | 1 | 0% | 2,673 | 2,908 | +9% | 0 | 0 | — |
case-16 | pass→pass | 16,363 | 11,955 | -27% | 1 | 1 | 0% | 2,591 | 2,385 | -8% | 0 | 0 | — |
case-17 | pass→pass | 7,775 | 6,840 | -12% | 1 | 1 | 0% | 1,382 | 1,471 | +6% | 0 | 0 | — |
case-18 | fail→pass | 13,158 | 10,368 | -21% | 1 | 1 | 0% | 2,041 | 2,178 | +7% | 0 | 0 | — |
case-19 | fail→pass | 21,182 | 18,641 | -12% | 1 | 1 | 0% | 3,576 | 3,545 | -1% | 0 | 0 | — |
case-20 | pass→pass | 20,023 | 18,644 | -7% | 1 | 1 | 0% | 3,306 | 3,114 | -6% | 0 | 0 | — |
case-21 | pass→pass | 19,459 | 19,019 | -2% | 1 | 1 | 0% | 3,329 | 3,713 | +12% | 0 | 0 | — |
case-22 | pass→pass | 19,014 | 16,668 | -12% | 1 | 1 | 0% | 3,569 | 3,565 | -0% | 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. 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.