Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply pre-composed aesthetic archetype defaults to jumpstart brand design work covering color, typography, layout, voice, and imagery direction. Use this skill whenever the user wants a starting point for brand visual and verbal direction, references a brand they want to design near (for example 'something like Stripe' or 'editorial like Linear'), needs to converge faster from many directions to one, or wants known-good defaults for a specific industry vertical. Triggers on archetype, design arc
.claude/skills/rampstackco-brand-archetype-system/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 108% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 97% | 0% |
Pre-composed aesthetic archetype defaults that jumpstart brand design work. Each archetype bundles color tendencies, type pairings, layout patterns, voice samples, and imagery direction for a recognizable aesthetic family. Stack-agnostic. Tool-agnostic.
This is a starting-point library, not a methodology library. The agent applies archetype defaults to a specific brief; the archetypes themselves provide anchor points, not endpoints.
brand-identity)brand-ideation)brand-voice)brand-style-guide)This skill works upstream of brand-identity (provides starting defaults that brand-identity refines into a finished system) and parallel to creative-direction (archetypes can be located by position on the 4 creative-direction axes).
Typical flow when this skill is invoked:
Direct verbatim copying of an archetype's default palette into a new brand is a failure mode. Archetypes constrain the design space; the brief specifies the point within that space.
Two layers of content in references/:
Located in references/core-archetypes/. Each is an aesthetic family with:
creative-direction axesLocated in references/by-vertical/. Each maps named brands to core archetypes in that vertical with:
00-archetype-system-overview.md: how the system works, the taxonomy01-how-to-apply-an-archetype.md: the adaptation discipline, the failure modes| # | Archetype | Single-line summary | |---|---|---| | 01 | Editorial Restrained | Low-saturation, type-led, generous whitespace, considered | | 02 | Technical Precise | Monospace and grid prominent, data-dense, system-feeling | | 03 | Warm Conversational | Human imagery, mid-saturation, friendly type, approachable | | 04 | Bold Confident | High-contrast, large display type, saturated, direct | | 05 | Playful Energetic | Bright colors, illustration-led, dynamic, character-driven | | 06 | Luxe Considered | Serif-led, generous spacing, restrained palette, premium | | 07 | Clinical Trustworthy | Cool palette, sans-serif, clean medical or financial register | | 08 | Rugged Utilitarian | Earth tones, workwear influence, no-nonsense type | | 09 | Retro Nostalgic | Period-specific palette and type, intentional vintage reference | | 10 | Minimal Essentialist | Black, white, single accent, sans-serif only, sparse | | 11 | Vibrant Saturated | High-saturation across full palette, color as character | | 12 | Documentary Honest | Photography-led, real people, low-touched imagery |
Archetypes are NAMED for aesthetic families, NOT for brands. "Editorial Restrained" not "Stripe-like." Brands are referenced as exemplars in description text using attribution language: "exemplified by brands]", "common among brands]", "characteristic of brands]".
This is descriptive and nominative fair use territory, durable across brand redesigns. A brand that pivots its identity does not invalidate the archetype it once exemplified.
Each archetype positions on the 4 axes defined in the creative-direction skill. Brief summary for cross-reference:
Each core archetype file specifies its position on each axis.
references/00-archetype-system-overview.md - How the two-layer system fits together, entry patterns, when archetypes compose, and the recurring failure modes.references/01-how-to-apply-an-archetype.md - The 5-step adaptation process: locate the design space, map the brief to adjustable dimensions, adapt color, adapt type, write voice samples. Includes the cross-archetype coherence check.The 12 core archetype files live in the core-archetypes/ subdirectory under references, numbered 01 (Editorial Restrained) through 12 (Documentary Honest). The 18 vertical application files live in the by-vertical/ subdirectory under references, numbered 01 (B2B SaaS Developer Tools) through 18 (Real Estate and Proptech). Load the relevant core archetype file plus the relevant vertical file together for any brief that names both an aesthetic family and an industry.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 21,298 | 21,360 | +0% | 1 | 1 | 0% | 3,463 | 5,055 | +46% | 0 | 0 | — |
case-02 | fail→pass | 23,992 | 22,254 | -7% | 1 | 1 | 0% | 3,902 | 5,330 | +37% | 0 | 0 | — |
case-03 | pass→pass | 28,444 | 19,697 | -31% | 1 | 1 | 0% | 4,437 | 4,566 | +3% | 0 | 0 | — |
case-04 | pass→pass | 17,581 | 18,076 | +3% | 1 | 1 | 0% | 2,732 | 4,575 | +67% | 0 | 0 | — |
case-05 | pass→pass | 19,588 | 19,239 | -2% | 1 | 1 | 0% | 3,171 | 4,497 | +42% | 0 | 0 | — |
case-06 | pass→pass | 20,004 | 15,237 | -24% | 1 | 1 | 0% | 3,870 | 4,341 | +12% | 0 | 0 | — |
case-07 | fail→fail | 21,071 | 18,969 | -10% | 1 | 1 | 0% | 3,298 | 4,588 | +39% | 0 | 0 | — |
case-08 | pass→pass | 12,281 | 10,733 | -13% | 1 | 1 | 0% | 1,906 | 3,215 | +69% | 0 | 0 | — |
case-09 | fail→pass | 12,107 | 9,599 | -21% | 1 | 1 | 0% | 1,818 | 3,066 | +69% | 0 | 0 | — |
case-10 | pass→fail | 17,286 | 14,044 | -19% | 1 | 1 | 0% | 2,470 | 3,604 | +46% | 0 | 0 | — |
case-11 | fail→fail | 18,331 | 23,817 | +30% | 1 | 1 | 0% | 2,711 | 5,234 | +93% | 0 | 0 | — |
case-12 | fail→pass | 20,851 | 21,565 | +3% | 1 | 1 | 0% | 3,231 | 4,807 | +49% | 0 | 0 | — |
case-13 | fail→pass | 15,799 | 22,192 | +40% | 1 | 1 | 0% | 2,231 | 4,645 | +108% | 0 | 0 | — |
case-14 | fail→pass | 18,248 | 22,222 | +22% | 1 | 1 | 0% | 2,469 | 4,856 | +97% | 0 | 0 | — |
case-15 | fail→pass | 22,939 | 26,133 | +14% | 1 | 1 | 0% | 3,953 | 5,901 | +49% | 0 | 0 | — |
case-16 | fail→pass | 23,503 | 22,250 | -5% | 1 | 1 | 0% | 3,810 | 5,226 | +37% | 0 | 0 | — |
case-17 | fail→fail | 20,602 | 22,823 | +11% | 1 | 1 | 0% | 3,163 | 5,195 | +64% | 0 | 0 | — |
case-18 | fail→pass | 19,803 | 22,490 | +14% | 1 | 1 | 0% | 2,900 | 5,127 | +77% | 0 | 0 | — |
case-19 | fail→pass | 23,155 | 21,941 | -5% | 1 | 1 | 0% | 3,491 | 5,085 | +46% | 0 | 0 | — |
case-20 | fail→pass | 21,549 | 25,217 | +17% | 1 | 1 | 0% | 3,277 | 5,490 | +68% | 0 | 0 | — |
case-21 | fail→pass | 20,651 | 14,812 | -28% | 1 | 1 | 0% | 3,250 | 4,024 | +24% | 0 | 0 | — |
case-22 | fail→fail | 16,108 | 18,515 | +15% | 1 | 1 | 0% | 2,371 | 4,406 | +86% | 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.