Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build or audit a comprehensive brand style guide that documents the full brand system including story, logo system, color, typography, imagery, voice, applications, and dos/don'ts. Use this skill whenever the user wants to create brand guidelines, document an existing brand, build a brand book, audit an existing style guide for completeness, or produce the artifact that other teams will reference for years. Triggers on style guide, brand guidelines, brand book, brand standards, brand manual, sty
.claude/skills/rampstackco-brand-style-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 125% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 30% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 121% | 0% |
Document the brand system so other people can use it without ambiguity. This is the artifact that lives longest. Designers, developers, agencies, and vendors will reference it for years. Build it like a reference manual, not a presentation.
This skill assumes the brand identity is designed (run brand-identity first if not). The output of this skill is the canonical reference document.
brand-identity)brand-voice for the voice doc, then integrate)design-standards or design-system)A complete style guide has eight sections. Most guides skip 2 or 3 of them and create downstream confusion. Build all 8 from the start.
The narrative behind the brand. Why it exists, what it stands for, what it rejects.
Every variant of every mark, with rules.
The full color system with rules.
The full type system.
What pictures look like in this brand.
How the brand sounds. (Pulled from brand-voice work if done separately.)
The brand applied to real contexts.
The boundaries, illustrated.
The dos and don'ts section is what people actually reference in practice. Make it the easiest section to scan.
references/style-guide-template.md.Default output is a multi-section markdown document or a structured set of files, plus a presentation-ready version (web page, PDF, or Figma) for sharing with stakeholders.
Recommended structure:
brand/
style-guide.md (the canonical document)
story.md (or as a section)
logo/
construction.md
files/ (SVG, PNG, etc.)
colors.md
typography.md
imagery/
photography.md
illustration.md
icons/
icons.md
files/
voice.md
applications/
web.md
email.md
social.md
print.md
dos-and-donts.mdFor consumer-facing presentation, build a web page version that imports from these source files. The source files are canonical. The presentation is a view of them.
references/style-guide-template.md - Fillable section-by-section template.references/maintenance-playbook.md - How to keep the guide current after launch.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 32,968 | 35,153 | +7% | 1 | 1 | 0% | 6,238 | 8,046 | +29% | 0 | 0 | — |
case-02 | fail→fail | 34,926 | 33,580 | -4% | 1 | 1 | 0% | 6,233 | 8,041 | +29% | 0 | 0 | — |
case-03 | fail→fail | 34,771 | 34,560 | -1% | 1 | 1 | 0% | 6,221 | 8,029 | +29% | 0 | 0 | — |
case-04 | pass→pass | 18,274 | 12,478 | -32% | 1 | 1 | 0% | 2,915 | 3,786 | +30% | 0 | 0 | — |
case-05 | pass→pass | 16,634 | 26,784 | +61% | 1 | 1 | 0% | 2,731 | 6,029 | +121% | 0 | 0 | — |
case-06 | pass→pass | 19,065 | 22,984 | +21% | 1 | 1 | 0% | 3,974 | 6,619 | +67% | 0 | 0 | — |
case-07 | fail→pass | 14,355 | 10,532 | -27% | 1 | 1 | 0% | 2,060 | 3,325 | +61% | 0 | 0 | — |
case-08 | fail→pass | 18,797 | 17,250 | -8% | 1 | 1 | 0% | 2,796 | 4,598 | +64% | 0 | 0 | — |
case-09 | pass→pass | 20,606 | 21,596 | +5% | 1 | 1 | 0% | 3,667 | 5,638 | +54% | 0 | 0 | — |
case-10 | pass→pass | 19,788 | 25,329 | +28% | 1 | 1 | 0% | 2,956 | 6,063 | +105% | 0 | 0 | — |
case-11 | pass→pass | 15,639 | 14,921 | -5% | 1 | 1 | 0% | 2,316 | 3,918 | +69% | 0 | 0 | — |
case-17 | fail→fail | 15,677 | 10,880 | -31% | 1 | 1 | 0% | 2,575 | 3,670 | +43% | 0 | 0 | — |
case-12 | pass→pass | 16,461 | 21,133 | +28% | 1 | 1 | 0% | 2,658 | 4,841 | +82% | 0 | 0 | — |
case-13 | fail→fail | 20,846 | 23,788 | +14% | 1 | 1 | 0% | 2,882 | 5,138 | +78% | 0 | 0 | — |
case-14 | pass→pass | 12,458 | 8,143 | -35% | 1 | 1 | 0% | 1,850 | 3,006 | +62% | 0 | 0 | — |
case-15 | pass→pass | 13,293 | 13,735 | +3% | 1 | 1 | 0% | 1,972 | 3,896 | +98% | 0 | 0 | — |
case-16 | fail→pass | 10,063 | 10,767 | +7% | 1 | 1 | 0% | 1,497 | 3,371 | +125% | 0 | 0 | — |
case-18 | pass→pass | 16,158 | 17,485 | +8% | 1 | 1 | 0% | 2,336 | 4,453 | +91% | 0 | 0 | — |
case-19 | pass→pass | 15,003 | 12,377 | -18% | 1 | 1 | 0% | 2,297 | 3,539 | +54% | 0 | 0 | — |
case-20 | fail→fail | 15,758 | 11,253 | -29% | 1 | 1 | 0% | 2,510 | 3,494 | +39% | 0 | 0 | — |
case-21 | pass→pass | 11,246 | 11,877 | +6% | 1 | 1 | 0% | 1,761 | 3,602 | +105% | 0 | 0 | — |
case-22 | pass→pass | 8,812 | 3,530 | -60% | 1 | 1 | 0% | 1,401 | 2,310 | +65% | 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 +14 percentage points is the difference between those two pass rates over the 22 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.