Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Visual philosophy and art-direction for frontend. Use when creating high-concept work, campaigns, or when the user asks for a visual philosophy, manifesto, or unmistakable art-like aesthetic.
.claude/skills/sickn33-design-philosophy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -17% | 0% |
Use this skill when you need visual philosophy and art-direction for frontend. Use when creating high-concept work, campaigns, or when the user asks for a visual philosophy, manifesto, or unmistakable art-like aesthetic.
Apply with design for high-concept work, campaigns, or when the user asks for a visual philosophy, manifesto, or art-like aesthetic. This skill guides creating a named movement and expressing it visually.
For philosophy examples, see reference.md. The numbered checklist above is the canonical generation procedure.
Before building: identify one subtle conceptual thread from the request. The topic is a subtle, niche reference embedded in the work — not literal, always sophisticated. Someone familiar with the subject should feel it intuitively; others experience a strong abstract composition. Weave it into form, color, and composition. "Jazz quote" principle: only those who know catch it; everyone benefits.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 22,165 | 14,633 | -34% | 1 | 1 | 0% | 3,457 | 2,933 | -15% | 0 | 0 | — |
case-03 | fail→pass | 25,777 | 11,206 | -57% | 1 | 1 | 0% | 3,967 | 2,313 | -42% | 0 | 0 | — |
case-22 | pass→pass | 14,872 | 12,399 | -17% | 1 | 1 | 0% | 3,991 | 3,456 | -13% | 0 | 0 | — |
case-01 | fail→pass | 19,952 | 11,104 | -44% | 1 | 1 | 0% | 3,286 | 2,377 | -28% | 0 | 0 | — |
case-04 | fail→pass | 24,787 | 10,421 | -58% | 1 | 1 | 0% | 3,848 | 2,146 | -44% | 0 | 0 | — |
case-05 | fail→pass | 20,551 | 13,047 | -37% | 1 | 1 | 0% | 3,135 | 2,616 | -17% | 0 | 0 | — |
case-06 | fail→pass | 18,163 | 10,918 | -40% | 1 | 1 | 0% | 2,949 | 2,033 | -31% | 0 | 0 | — |
case-07 | fail→pass | 18,521 | 12,974 | -30% | 1 | 1 | 0% | 2,598 | 2,116 | -19% | 0 | 0 | — |
case-08 | fail→fail | 20,948 | 13,179 | -37% | 1 | 1 | 0% | 3,055 | 2,235 | -27% | 0 | 0 | — |
case-09 | pass→pass | 19,941 | 10,397 | -48% | 1 | 1 | 0% | 3,053 | 1,983 | -35% | 0 | 0 | — |
case-10 | fail→pass | 19,941 | 10,865 | -46% | 1 | 1 | 0% | 2,892 | 1,959 | -32% | 0 | 0 | — |
case-11 | fail→pass | 19,969 | 13,076 | -35% | 1 | 1 | 0% | 3,003 | 2,405 | -20% | 0 | 0 | — |
case-12 | fail→pass | 20,335 | 10,310 | -49% | 1 | 1 | 0% | 2,858 | 1,952 | -32% | 0 | 0 | — |
case-13 | pass→fail | 17,231 | 16,229 | -6% | 1 | 1 | 0% | 2,623 | 2,813 | +7% | 0 | 0 | — |
case-14 | fail→pass | 12,841 | 13,063 | +2% | 1 | 1 | 0% | 1,960 | 2,365 | +21% | 0 | 0 | — |
case-15 | fail→fail | 15,780 | 12,248 | -22% | 1 | 1 | 0% | 2,301 | 2,271 | -1% | 0 | 0 | — |
case-16 | fail→fail | 16,554 | 12,920 | -22% | 1 | 1 | 0% | 2,496 | 2,462 | -1% | 0 | 0 | — |
case-17 | fail→fail | 22,442 | 12,255 | -45% | 1 | 1 | 0% | 3,447 | 2,256 | -35% | 0 | 0 | — |
case-18 | pass→pass | 22,340 | 12,025 | -46% | 1 | 1 | 0% | 3,192 | 2,544 | -20% | 0 | 0 | — |
case-19 | fail→fail | 17,357 | 12,055 | -31% | 1 | 1 | 0% | 2,773 | 2,252 | -19% | 0 | 0 | — |
case-20 | pass→pass | 9,515 | 6,670 | -30% | 1 | 1 | 0% | 2,128 | 1,998 | -6% | 0 | 0 | — |
case-21 | pass→pass | 11,773 | 12,961 | +10% | 1 | 1 | 0% | 2,371 | 3,496 | +47% | 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 +45 percentage points is the difference between those two pass rates over the 22 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.