Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
.claude/skills/aifinlab-brand-guidelines/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -47% | 0% |
To access Anthropic's official brand identity and style resources, use this skill.
Keywords: branding, corporate identity, visual identity, post-processing, styling, brand colors, typography, Anthropic brand, visual formatting, visual design
Main Colors:
#141413 - Primary text and dark backgrounds#faf9f5 - Light backgrounds and text on dark#b0aea5 - Secondary elements#e8e6dc - Subtle backgroundsAccent Colors:
#d97757 - Primary accent#6a9bcc - Secondary accent#788c5d - Tertiary accentpython# 调用 skill result = run_skill({ "param1": "value1", "param2": "value2" })
bashpython scripts/run_skill.py --input data.json
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 65,948 | 55,459 | -16% | 1 | 1 | 0% | 6,591 | 5,844 | -11% | 0 | 0 | — |
case-02 | fail→pass | 24,324 | 14,349 | -41% | 1 | 1 | 0% | 4,453 | 3,715 | -17% | 0 | 0 | — |
case-03 | fail→pass | 28,600 | 22,060 | -23% | 1 | 1 | 0% | 5,325 | 5,015 | -6% | 0 | 0 | — |
case-04 | fail→pass | 8,472 | 3,423 | -60% | 1 | 1 | 0% | 1,297 | 1,119 | -14% | 0 | 0 | — |
case-05 | fail→pass | 40,927 | 22,323 | -45% | 1 | 1 | 0% | 1,705 | 908 | -47% | 0 | 0 | — |
case-06 | fail→pass | 40,886 | 2,068 | -95% | 1 | 1 | 0% | 1,529 | 804 | -47% | 0 | 0 | — |
case-07 | pass→pass | 8,763 | 2,019 | -77% | 1 | 1 | 0% | 1,236 | 829 | -33% | 0 | 0 | — |
case-08 | fail→pass | 49,823 | 2,082 | -96% | 1 | 1 | 0% | 3,325 | 858 | -74% | 0 | 0 | — |
case-09 | fail→pass | 11,219 | 2,519 | -78% | 1 | 1 | 0% | 1,779 | 900 | -49% | 0 | 0 | — |
case-10 | fail→pass | 20,187 | 1,637 | -92% | 1 | 1 | 0% | 3,447 | 746 | -78% | 0 | 0 | — |
case-11 | fail→pass | 13,948 | 2,292 | -84% | 1 | 1 | 0% | 2,194 | 869 | -60% | 0 | 0 | — |
case-12 | pass→pass | 40,710 | 1,991 | -95% | 1 | 1 | 0% | 1,596 | 839 | -47% | 0 | 0 | — |
case-13 | fail→pass | 31,563 | 1,999 | -94% | 1 | 1 | 0% | 5,268 | 833 | -84% | 0 | 0 | — |
case-14 | fail→pass | 24,329 | 2,012 | -92% | 1 | 1 | 0% | 3,842 | 823 | -79% | 0 | 0 | — |
case-15 | pass→pass | 3,753 | 3,179 | -15% | 1 | 1 | 0% | 712 | 1,049 | +47% | 0 | 0 | — |
case-16 | pass→pass | 16,269 | 6,076 | -63% | 1 | 1 | 0% | 2,328 | 1,570 | -33% | 0 | 0 | — |
case-17 | fail→pass | 14,656 | 1,735 | -88% | 1 | 1 | 0% | 2,266 | 773 | -66% | 0 | 0 | — |
case-18 | pass→pass | 17,127 | 5,081 | -70% | 1 | 1 | 0% | 2,468 | 1,346 | -45% | 0 | 0 | — |
case-19 | fail→pass | 11,759 | 2,365 | -80% | 1 | 1 | 0% | 1,893 | 944 | -50% | 0 | 0 | — |
case-20 | fail→pass | 13,789 | 31,363 | +127% | 1 | 1 | 0% | 1,973 | 2,217 | +12% | 0 | 0 | — |
case-21 | pass→pass | 7,929 | 6,095 | -23% | 1 | 1 | 0% | 1,092 | 1,369 | +25% | 0 | 0 | — |
case-22 | pass→pass | 19,353 | 18,432 | -5% | 1 | 1 | 0% | 3,256 | 3,749 | +15% | 0 | 0 | — |
case-23 | pass→pass | 17,839 | 15,489 | -13% | 1 | 1 | 0% | 3,346 | 3,547 | +6% | 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 +65 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.