Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply professional font and color themes to artifacts including slides, docs, reports, and HTML landing pages. Ships 10 pre-set themes.
.claude/skills/nexu-io-theme-factory/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 92% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-13 | ✓→✗ | ▼ Worse | 25% | 0% |
> Curated from Anthropic's official skills repository.
Apply professional font and color themes to artifacts including slides, docs, reports, and HTML landing pages. Ships 10 pre-set themes.
design-systemsThis catalogue entry advertises the skill in OpenDesign so the agent discovers it during planning. To run the full upstream workflow with its original assets, scripts, and references, install the upstream bundle into your active agent's skills directory:
bash# Inspect the upstream README for exact paths open https://github.com/anthropics/skills/tree/main/skills/theme-factory
Then ask the agent to invoke this skill by name (theme-factory) or with one of the trigger phrases listed in this skill's frontmatter.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 36,494 | 54,939 | +51% | 1 | 1 | 0% | 8,227 | 8,199 | -0% | 0 | 0 | — |
case-02 | pass→pass | 49,025 | 30,784 | -37% | 1 | 1 | 0% | 8,223 | 5,213 | -37% | 0 | 0 | — |
case-03 | fail→fail | 10,753 | 25,673 | +139% | 1 | 1 | 0% | 1,766 | 5,417 | +207% | 0 | 0 | — |
case-04 | pass→pass | 18,396 | 26,902 | +46% | 1 | 1 | 0% | 3,373 | 4,166 | +24% | 0 | 0 | — |
case-05 | pass→pass | 34,755 | 154,152 | +344% | 1 | 1 | 0% | 5,212 | 7,023 | +35% | 0 | 0 | — |
case-06 | pass→pass | 38,273 | 21,474 | -44% | 1 | 1 | 0% | 4,547 | 4,841 | +6% | 0 | 0 | — |
case-07 | fail→pass | 27,876 | 28,736 | +3% | 1 | 1 | 0% | 4,340 | 4,822 | +11% | 0 | 0 | — |
case-08 | fail→pass | 21,965 | 25,857 | +18% | 1 | 1 | 0% | 2,943 | 5,663 | +92% | 0 | 0 | — |
case-09 | pass→pass | 22,621 | 37,004 | +64% | 1 | 1 | 0% | 4,286 | 5,890 | +37% | 0 | 0 | — |
case-10 | fail→pass | 17,092 | 24,004 | +40% | 1 | 1 | 0% | 3,155 | 3,895 | +23% | 0 | 0 | — |
case-11 | pass→pass | 31,341 | 21,542 | -31% | 1 | 1 | 0% | 3,652 | 3,502 | -4% | 0 | 0 | — |
case-12 | fail→fail | 33,699 | 22,946 | -32% | 1 | 1 | 0% | 7,138 | 4,780 | -33% | 0 | 0 | — |
case-13 | pass→fail | 16,293 | 25,527 | +57% | 1 | 1 | 0% | 3,410 | 4,247 | +25% | 0 | 0 | — |
case-14 | pass→pass | 19,353 | 17,556 | -9% | 1 | 1 | 0% | 4,037 | 4,136 | +2% | 0 | 0 | — |
case-15 | pass→pass | 17,888 | 23,991 | +34% | 1 | 1 | 0% | 3,544 | 5,348 | +51% | 0 | 0 | — |
case-16 | pass→pass | 32,000 | 29,618 | -7% | 1 | 1 | 0% | 5,050 | 4,811 | -5% | 0 | 0 | — |
case-17 | pass→pass | 22,524 | 52,412 | +133% | 1 | 1 | 0% | 4,607 | 4,497 | -2% | 0 | 0 | — |
case-18 | fail→pass | 19,188 | 28,486 | +48% | 1 | 1 | 0% | 2,993 | 4,463 | +49% | 0 | 0 | — |
case-19 | fail→fail | 51,183 | 50,249 | -2% | 1 | 1 | 0% | 8,228 | 8,444 | +3% | 0 | 0 | — |
case-20 | pass→pass | 18,900 | 24,279 | +28% | 1 | 1 | 0% | 2,480 | 3,410 | +38% | 0 | 0 | — |
case-21 | pass→pass | 15,106 | 12,092 | -20% | 1 | 1 | 0% | 2,284 | 2,102 | -8% | 0 | 0 | — |
case-22 | pass→pass | 16,593 | 19,979 | +20% | 1 | 1 | 0% | 3,180 | 3,374 | +6% | 0 | 0 | — |
case-23 | pass→pass | 32,465 | 30,272 | -7% | 1 | 1 | 0% | 5,150 | 6,446 | +25% | 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 +13 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is 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.