Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when creating or refining WebGL-heavy landing pages and you need to steer toward a specific visual outcome (premium, technical, playful, cinematic) while balancing conversion clarity, performance, and implementation complexity.
.claude/skills/mengto-webgl-landing-steering/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 31% | 0% |
Map landing-page goal to WebGL direction before writing code.
Identify the primary conversion and brand signal:
Also capture:
prefers-reduced-motion policy)Pick one dominant lane; avoid mixing 3-4 heavy effects in the hero.
Use for: SaaS, productivity, B2B tools where readability wins.
Use for: AI, infra, analytics, developer products.
Use for: hardware, apps with strong product visuals, launches.
Use for: campaign pages where wow factor is the main KPI.
Pass these gates before adding more visual complexity:
Math.min(devicePixelRatio, 1.5-2).prefers-reduced-motion (still frame or low-motion mode).Default to low/medium risk for conversion pages unless user explicitly asks for campaign-grade immersion.
Use this prompt pattern when asked to build a WebGL landing hero:
"Build a lane] WebGL hero for a page type] with brand adjectives]. Primary goal: conversion]. Constraints: device mix], performance budget], reduced motion policy]. Implement fallback first, then enhance with WebGL. Keep hero copy clarity as priority over visual complexity."
Return:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | pass→pass | 15,910 | 15,993 | +1% | 1 | 1 | 0% | 3,062 | 4,453 | +45% | 0 | 0 | — |
case-01 | fail→pass | 34,383 | 15,903 | -54% | 1 | 1 | 0% | 6,254 | 4,002 | -36% | 0 | 0 | — |
case-02 | fail→pass | 26,878 | 15,842 | -41% | 1 | 1 | 0% | 4,842 | 3,926 | -19% | 0 | 0 | — |
case-03 | fail→pass | 15,976 | 14,995 | -6% | 1 | 1 | 0% | 2,582 | 3,645 | +41% | 0 | 0 | — |
case-04 | fail→pass | 15,436 | 13,157 | -15% | 1 | 1 | 0% | 2,618 | 3,287 | +26% | 0 | 0 | — |
case-05 | fail→fail | 17,388 | 12,033 | -31% | 1 | 1 | 0% | 3,032 | 3,068 | +1% | 0 | 0 | — |
case-06 | fail→pass | 16,743 | 13,737 | -18% | 1 | 1 | 0% | 2,809 | 3,678 | +31% | 0 | 0 | — |
case-08 | pass→pass | 12,937 | 10,552 | -18% | 1 | 1 | 0% | 2,347 | 3,099 | +32% | 0 | 0 | — |
case-09 | pass→pass | 14,080 | 14,460 | +3% | 1 | 1 | 0% | 2,745 | 4,111 | +50% | 0 | 0 | — |
case-10 | fail→pass | 16,048 | 13,277 | -17% | 1 | 1 | 0% | 2,344 | 3,411 | +46% | 0 | 0 | — |
case-11 | fail→pass | 14,280 | 12,863 | -10% | 1 | 1 | 0% | 2,541 | 3,634 | +43% | 0 | 0 | — |
case-12 | pass→pass | 13,246 | 11,299 | -15% | 1 | 1 | 0% | 2,392 | 2,986 | +25% | 0 | 0 | — |
case-13 | pass→pass | 15,762 | 15,595 | -1% | 1 | 1 | 0% | 2,863 | 4,061 | +42% | 0 | 0 | — |
case-14 | fail→pass | 15,569 | 11,498 | -26% | 1 | 1 | 0% | 2,595 | 3,047 | +17% | 0 | 0 | — |
case-15 | fail→fail | 13,213 | 10,336 | -22% | 1 | 1 | 0% | 2,183 | 2,921 | +34% | 0 | 0 | — |
case-16 | fail→pass | 12,615 | 9,786 | -22% | 1 | 1 | 0% | 2,146 | 2,683 | +25% | 0 | 0 | — |
case-17 | fail→pass | 16,262 | 16,786 | +3% | 1 | 1 | 0% | 3,179 | 4,483 | +41% | 0 | 0 | — |
case-18 | pass→pass | 14,023 | 15,541 | +11% | 1 | 1 | 0% | 2,371 | 3,737 | +58% | 0 | 0 | — |
case-19 | pass→pass | 11,762 | 10,437 | -11% | 1 | 1 | 0% | 2,025 | 2,742 | +35% | 0 | 0 | — |
case-20 | fail→pass | 17,010 | 7,764 | -54% | 1 | 1 | 0% | 3,040 | 2,448 | -19% | 0 | 0 | — |
case-21 | pass→pass | 11,558 | 15,063 | +30% | 1 | 1 | 0% | 2,454 | 4,254 | +73% | 0 | 0 | — |
case-22 | pass→pass | 17,422 | 15,270 | -12% | 1 | 1 | 0% | 3,110 | 3,941 | +27% | 0 | 0 | — |
case-23 | pass→pass | 11,400 | 11,521 | +1% | 1 | 1 | 0% | 2,400 | 3,509 | +46% | 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 +48 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.