Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Define a product's motion design language -- duration and easing scales, choreography rules, and signature moves -- and export it as tokens plus ready-to-use Framer Motion variants and CSS. The difference between animations and a motion system: everything moves like it belongs to the same product.
.claude/skills/onewave-ai-motion-language-designer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 32% | 0% |
Design motion the way type and color get designed: as a system with a scale, roles, and rules -- not per-component improvisation. Output is a motion spec plus drop-in implementation (Framer Motion variants, CSS custom properties and keyframes) so the system is enforceable, not aspirational.
instant ~100ms for state feedback, fast ~180-220ms for micro-interactions, base ~300ms for reveals, slow ~500ms+ for scene changes) and an easing vocabulary (standard, decelerate for entrances, accelerate for exits, plus at most ONE signature spring/overshoot -- the personality lives here). Calibrate values to the brand read from step 1, not generic defaults.prefers-reduced-motion behavior: fade or nothing).motion-tokens.json (durations, easings, distances), motion.css (custom properties + keyframes + reduced-motion guards), motion.ts (Framer Motion variants for each signature move, referencing the tokens), and MOTION.md -- the spec a new developer reads before animating anything.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 31,903 | 42,131 | +32% | 1 | 1 | 0% | 6,206 | 6,756 | +9% | 0 | 0 | — |
case-02 | fail→pass | 21,518 | 29,496 | +37% | 1 | 1 | 0% | 4,713 | 6,986 | +48% | 0 | 0 | — |
case-03 | fail→fail | 26,094 | 29,105 | +12% | 1 | 1 | 0% | 6,200 | 6,215 | +0% | 0 | 0 | — |
case-04 | pass→fail | 28,587 | 24,053 | -16% | 1 | 1 | 0% | 6,186 | 5,885 | -5% | 0 | 0 | — |
case-05 | pass→pass | 41,100 | 27,222 | -34% | 1 | 1 | 0% | 4,306 | 6,010 | +40% | 0 | 0 | — |
case-06 | pass→fail | 25,538 | 23,563 | -8% | 1 | 1 | 0% | 6,188 | 5,929 | -4% | 0 | 0 | — |
case-07 | fail→fail | 29,328 | 17,831 | -39% | 1 | 1 | 0% | 5,197 | 4,741 | -9% | 0 | 0 | — |
case-08 | fail→pass | 15,979 | 17,679 | +11% | 1 | 1 | 0% | 3,515 | 4,729 | +35% | 0 | 0 | — |
case-09 | fail→fail | 14,730 | 17,320 | +18% | 1 | 1 | 0% | 2,971 | 4,462 | +50% | 0 | 0 | — |
case-10 | fail→fail | 12,186 | 14,704 | +21% | 1 | 1 | 0% | 2,503 | 3,895 | +56% | 0 | 0 | — |
case-11 | fail→pass | 9,508 | 11,191 | +18% | 1 | 1 | 0% | 2,224 | 3,235 | +45% | 0 | 0 | — |
case-12 | fail→fail | 17,148 | 18,306 | +7% | 1 | 1 | 0% | 3,188 | 4,832 | +52% | 0 | 0 | — |
case-13 | fail→pass | 15,294 | 13,414 | -12% | 1 | 1 | 0% | 2,592 | 3,291 | +27% | 0 | 0 | — |
case-14 | fail→fail | 22,103 | 30,328 | +37% | 1 | 1 | 0% | 5,053 | 5,865 | +16% | 0 | 0 | — |
case-15 | fail→pass | 17,243 | 16,224 | -6% | 1 | 1 | 0% | 3,140 | 4,146 | +32% | 0 | 0 | — |
case-16 | fail→fail | 27,791 | 26,662 | -4% | 1 | 1 | 0% | 5,983 | 6,891 | +15% | 0 | 0 | — |
case-17 | fail→pass | 20,044 | 19,718 | -2% | 1 | 1 | 0% | 4,038 | 4,925 | +22% | 0 | 0 | — |
case-18 | fail→fail | 17,118 | 15,783 | -8% | 1 | 1 | 0% | 3,217 | 4,073 | +27% | 0 | 0 | — |
case-19 | fail→fail | 7,457 | 9,980 | +34% | 1 | 1 | 0% | 1,479 | 2,203 | +49% | 0 | 0 | — |
case-20 | fail→fail | 6,416 | 11,249 | +75% | 1 | 1 | 0% | 1,502 | 3,304 | +120% | 0 | 0 | — |
case-21 | fail→fail | 10,705 | 19,351 | +81% | 1 | 1 | 0% | 2,326 | 4,851 | +109% | 0 | 0 | — |
case-22 | fail→fail | 15,662 | 15,827 | +1% | 1 | 1 | 0% | 3,177 | 4,181 | +32% | 0 | 0 | — |
case-23 | fail→pass | 15,189 | 20,930 | +38% | 1 | 1 | 0% | 3,095 | 5,065 | +64% | 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 +22 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are 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.