Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Best practices and comprehensive guide for Remotion - programmatic video creation in React with animations, compositions, and media handling
.claude/skills/davila7-remotion/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 169% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -21% | 0% |
Comprehensive skill set for creating programmatic videos using Remotion, a framework for creating videos programmatically using React.
Use this skill whenever you are dealing with Remotion code to obtain domain-specific knowledge about:
Remotion allows you to create videos using:
useCurrentFrame() hookRead individual rule files for detailed explanations and code examples:
tsximport { useCurrentFrame, useVideoConfig, interpolate } from "remotion"; export const MyComposition = () => { const frame = useCurrentFrame(); const { fps } = useVideoConfig(); const opacity = interpolate(frame, [0, 2 * fps], [0, 1], { extrapolateRight: 'clamp', }); return ( <div style={{ opacity }}> <h1>Hello Remotion!</h1> </div> ); };
useCurrentFrame() - Drive all animations from the current framefps for frame calculations| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 17,343 | 14,002 | -19% | 1 | 1 | 0% | 3,516 | 4,304 | +22% | 0 | 0 | — |
case-02 | pass→pass | 15,031 | 9,785 | -35% | 1 | 1 | 0% | 2,539 | 3,190 | +26% | 0 | 0 | — |
case-03 | pass→pass | 9,393 | 6,441 | -31% | 1 | 1 | 0% | 1,791 | 2,538 | +42% | 0 | 0 | — |
case-04 | fail→pass | 11,010 | 4,367 | -60% | 1 | 1 | 0% | 2,047 | 2,205 | +8% | 0 | 0 | — |
case-05 | fail→pass | 6,769 | 3,690 | -45% | 1 | 1 | 0% | 1,175 | 2,068 | +76% | 0 | 0 | — |
case-22 | pass→pass | 11,892 | 8,000 | -33% | 1 | 1 | 0% | 2,295 | 2,923 | +27% | 0 | 0 | — |
case-06 | pass→pass | 10,256 | 5,383 | -48% | 1 | 1 | 0% | 2,012 | 2,404 | +19% | 0 | 0 | — |
case-07 | pass→pass | 7,338 | 5,315 | -28% | 1 | 1 | 0% | 1,214 | 2,327 | +92% | 0 | 0 | — |
case-08 | pass→pass | 7,165 | 5,166 | -28% | 1 | 1 | 0% | 1,260 | 2,289 | +82% | 0 | 0 | — |
case-09 | pass→pass | 17,462 | 9,777 | -44% | 1 | 1 | 0% | 3,320 | 3,288 | -1% | 0 | 0 | — |
case-10 | pass→pass | 10,074 | 6,309 | -37% | 1 | 1 | 0% | 1,746 | 2,499 | +43% | 0 | 0 | — |
case-11 | pass→pass | 16,477 | 10,954 | -34% | 1 | 1 | 0% | 2,953 | 3,443 | +17% | 0 | 0 | — |
case-12 | pass→pass | 10,667 | 10,487 | -2% | 1 | 1 | 0% | 2,205 | 3,522 | +60% | 0 | 0 | — |
case-13 | fail→pass | 15,235 | 5,721 | -62% | 1 | 1 | 0% | 2,941 | 2,473 | -16% | 0 | 0 | — |
case-14 | pass→pass | 10,838 | 6,147 | -43% | 1 | 1 | 0% | 1,956 | 2,412 | +23% | 0 | 0 | — |
case-15 | pass→pass | 13,901 | 7,676 | -45% | 1 | 1 | 0% | 2,036 | 2,802 | +38% | 0 | 0 | — |
case-16 | pass→pass | 10,555 | 4,750 | -55% | 1 | 1 | 0% | 1,869 | 2,260 | +21% | 0 | 0 | — |
case-17 | fail→pass | 4,696 | 3,820 | -19% | 1 | 1 | 0% | 776 | 2,086 | +169% | 0 | 0 | — |
case-18 | pass→pass | 15,624 | 6,284 | -60% | 1 | 1 | 0% | 2,023 | 2,465 | +22% | 0 | 0 | — |
case-19 | fail→pass | 17,986 | 7,697 | -57% | 1 | 1 | 0% | 3,501 | 2,781 | -21% | 0 | 0 | — |
case-20 | pass→pass | 5,638 | 4,809 | -15% | 1 | 1 | 0% | 985 | 2,350 | +139% | 0 | 0 | — |
case-21 | pass→pass | 16,382 | 13,353 | -18% | 1 | 1 | 0% | 3,132 | 4,168 | +33% | 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 +23 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.