Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this plugin when the user wants a "Data Rollup Frame" HyperFrames motion video — A native Remotion data frame — bars grow from zero by real data via spring physics while the figures roll 0→target in sync.
.claude/skills/nexu-io-video-template-frame-data-rollup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -38% | 0% |
A native Remotion data frame — bars grow from zero by real data via spring physics while the figures roll 0→target in sync.
A HyperFrames-ready Remotion (React/TSX) motion composition, bundled under source/. It renders deterministically to MP4 / WEBM at 16:9, 9:16, 1:1, default 15s, 30fps.
Best for: A data frame where the numbers should animate, not sit static · Weekly metrics / growth bars driven by real values · Enhancing one data segment of an otherwise hyperframes video
source/entry.ts to understand the named layers and the animation timeline.source/; do not introduce external network assets that would break a headless render.Source: html-video templates/frame-data-rollup (license Apache-2.0).
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→fail | 13,615 | 9,772 | -28% | 1 | 1 | 0% | 2,027 | 1,784 | -12% | 0 | 0 | — |
case-01 | fail→fail | 28,194 | 24,734 | -12% | 1 | 1 | 0% | 6,072 | 5,342 | -12% | 0 | 0 | — |
case-02 | fail→pass | 18,118 | 22,000 | +21% | 1 | 1 | 0% | 3,886 | 4,820 | +24% | 0 | 0 | — |
case-03 | fail→fail | 34,438 | 25,732 | -25% | 1 | 1 | 0% | 3,628 | 5,627 | +55% | 0 | 0 | — |
case-04 | fail→pass | 9,905 | 4,426 | -55% | 1 | 1 | 0% | 1,845 | 1,039 | -44% | 0 | 0 | — |
case-05 | fail→fail | 13,615 | 4,911 | -64% | 1 | 1 | 0% | 2,035 | 1,096 | -46% | 0 | 0 | — |
case-06 | fail→fail | 9,360 | 5,375 | -43% | 1 | 1 | 0% | 1,591 | 1,274 | -20% | 0 | 0 | — |
case-08 | pass→pass | 13,473 | 1,702 | -87% | 1 | 1 | 0% | 2,161 | 520 | -76% | 0 | 0 | — |
case-09 | fail→pass | 6,842 | 2,159 | -68% | 1 | 1 | 0% | 1,085 | 671 | -38% | 0 | 0 | — |
case-10 | pass→pass | 13,820 | 1,827 | -87% | 1 | 1 | 0% | 2,256 | 498 | -78% | 0 | 0 | — |
case-11 | fail→pass | 14,153 | 8,269 | -42% | 1 | 1 | 0% | 2,126 | 1,415 | -33% | 0 | 0 | — |
case-12 | pass→pass | 13,516 | 4,850 | -64% | 1 | 1 | 0% | 2,091 | 1,152 | -45% | 0 | 0 | — |
case-13 | fail→pass | 14,332 | 7,159 | -50% | 1 | 1 | 0% | 2,355 | 1,455 | -38% | 0 | 0 | — |
case-14 | fail→pass | 15,696 | 1,944 | -88% | 1 | 1 | 0% | 2,416 | 535 | -78% | 0 | 0 | — |
case-15 | fail→pass | 13,998 | 10,757 | -23% | 1 | 1 | 0% | 2,423 | 2,235 | -8% | 0 | 0 | — |
case-16 | pass→pass | 9,168 | 1,797 | -80% | 1 | 1 | 0% | 1,454 | 544 | -63% | 0 | 0 | — |
case-17 | pass→pass | 12,430 | 8,851 | -29% | 1 | 1 | 0% | 1,761 | 1,554 | -12% | 0 | 0 | — |
case-18 | pass→pass | 24,427 | 19,829 | -19% | 1 | 1 | 0% | 4,010 | 3,665 | -9% | 0 | 0 | — |
case-19 | pass→pass | 22,423 | 25,102 | +12% | 1 | 1 | 0% | 4,379 | 5,174 | +18% | 0 | 0 | — |
case-20 | pass→pass | 20,089 | 19,800 | -1% | 1 | 1 | 0% | 4,227 | 4,343 | +3% | 0 | 0 | — |
case-21 | pass→pass | 14,065 | 13,021 | -7% | 1 | 1 | 0% | 2,562 | 2,569 | +0% | 0 | 0 | — |
case-22 | fail→pass | 12,394 | 10,910 | -12% | 1 | 1 | 0% | 1,906 | 1,842 | -3% | 0 | 0 | — |
case-23 | pass→pass | 14,151 | 9,371 | -34% | 1 | 1 | 0% | 2,033 | 1,696 | -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. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 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.