Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate, preview, save, or validate Preline theme CSS using the bundled local generator in this skill. Use when the user asks to create a new Preline theme, match a brand color, translate a mood into a theme, generate light and dark theme CSS, preview theme output without writing files, or review and fix generated theme tokens. Do not use for generic non-Preline CSS work or arbitrary command execution.
.claude/skills/htmlstreamofficial-preline-theme-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -52% | 0% |
Use the bundled local scripts. Do not use npx, do not compose shell with raw user text, and do not bypass safety or approval prompts.
docs/workflow.md.docs/palette-guidance.md.docs/final-output-style.md before writing a final theme file.docs/validation-checklist.md before closing the task.examples.md only if the user wants stylistic guidance or sample outputs.docs/token-reference.md only if you need to inspect token coverage in detail.Follow docs/workflow.md for the full operating path.
Key requirements:
scripts/find-themes-dir.js.scripts/run-theme-generator.js.docs/validation-checklist.md.npx or any network-fetched package for this skill.find . traversal for path discovery; use scripts/find-themes-dir.js or a user-confirmed path.scripts/run-theme-generator.js as the only execution entry point for generation.docs/workflow.mddocs/palette-guidance.mddocs/final-output-style.mddocs/validation-checklist.mdexamples.mddocs/token-reference.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→pass | 9,547 | 7,299 | -24% | 1 | 1 | 0% | 1,816 | 1,980 | +9% | 0 | 0 | — |
case-01 | fail→fail | 18,023 | 4,400 | -76% | 1 | 1 | 0% | 4,125 | 789 | -81% | 0 | 0 | — |
case-20 | pass→pass | 12,089 | 12,347 | +2% | 1 | 1 | 0% | 2,719 | 3,583 | +32% | 0 | 0 | — |
case-02 | fail→fail | 18,760 | 5,748 | -69% | 1 | 1 | 0% | 4,222 | 764 | -82% | 0 | 0 | — |
case-03 | fail→fail | 2,328 | 4,327 | +86% | 1 | 1 | 0% | 370 | 756 | +104% | 0 | 0 | — |
case-04 | fail→pass | 9,555 | 2,922 | -69% | 1 | 1 | 0% | 1,734 | 1,069 | -38% | 0 | 0 | — |
case-05 | fail→pass | 7,526 | 6,071 | -19% | 1 | 1 | 0% | 1,523 | 1,185 | -22% | 0 | 0 | — |
case-06 | fail→pass | 5,942 | 1,495 | -75% | 1 | 1 | 0% | 1,029 | 768 | -25% | 0 | 0 | — |
case-07 | fail→fail | 8,370 | 1,904 | -77% | 1 | 1 | 0% | 1,470 | 815 | -45% | 0 | 0 | — |
case-08 | pass→pass | 2,609 | 2,561 | -2% | 1 | 1 | 0% | 411 | 864 | +110% | 0 | 0 | — |
case-09 | fail→pass | 7,173 | 1,484 | -79% | 1 | 1 | 0% | 1,122 | 821 | -27% | 0 | 0 | — |
case-10 | fail→pass | 7,764 | 1,648 | -79% | 1 | 1 | 0% | 1,479 | 714 | -52% | 0 | 0 | — |
case-21 | pass→pass | 6,033 | 5,342 | -11% | 1 | 1 | 0% | 1,307 | 1,594 | +22% | 0 | 0 | — |
case-11 | pass→pass | 10,686 | 4,937 | -54% | 1 | 1 | 0% | 2,066 | 1,377 | -33% | 0 | 0 | — |
case-12 | pass→pass | 11,711 | 3,005 | -74% | 1 | 1 | 0% | 2,100 | 1,091 | -48% | 0 | 0 | — |
case-13 | pass→pass | 7,975 | 3,022 | -62% | 1 | 1 | 0% | 1,609 | 935 | -42% | 0 | 0 | — |
case-14 | fail→pass | 6,829 | 1,618 | -76% | 1 | 1 | 0% | 1,215 | 850 | -30% | 0 | 0 | — |
case-15 | fail→pass | 10,973 | 2,711 | -75% | 1 | 1 | 0% | 2,315 | 951 | -59% | 0 | 0 | — |
case-16 | fail→fail | 11,491 | 5,006 | -56% | 1 | 1 | 0% | 2,478 | 781 | -68% | 0 | 0 | — |
case-17 | fail→fail | 24,071 | 5,179 | -78% | 1 | 1 | 0% | 6,165 | 935 | -85% | 0 | 0 | — |
case-18 | fail→fail | 6,493 | 1,562 | -76% | 1 | 1 | 0% | 1,146 | 804 | -30% | 0 | 0 | — |
case-19 | fail→fail | 14,210 | 5,015 | -65% | 1 | 1 | 0% | 2,649 | 1,548 | -42% | 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, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +32 percentage points is the difference between those two pass rates over the 17 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.