vertical-site-conventions
정보
이 스킬은 개발자가 산업별 디자인 관례에 맞춰 웹사이트를 제작하거나 수정하여, 전자상거래나 호텔업과 같은 특정 업종 내에서 신뢰할 수 있는 외관을 갖추도록 돕습니다. 사이트가 레이아웃, 구성 또는 상품 진열 문제로 인해 일반적이거나 해당 카테고리와 어울리지 않는다고 느껴질 때 발동됩니다. 이를 사용하여 기대되는 페이지 구조를 구현하고 경쟁사가 설정한 경험 기준을 충족하세요.
빠른 설치
Claude Code
추천npx skills add rampstackco/claude-skills -a claude-code/plugin add https://github.com/rampstackco/claude-skillsgit clone https://github.com/rampstackco/claude-skills.git ~/.claude/skills/vertical-site-conventionsClaude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요
문서
Vertical Site Conventions
Compose pages and whole sites to the layout, density, and merchandising conventions that make a build read as a credible member of its vertical. Stack-agnostic. Works on any framework that can express server-rendered pages and component composition.
This skill owns vertical-level composition: which modules go where, how dense the page is, how much of the catalog or offering is surfaced, and which paths the navigation serves. It routes the day-to-day visual decisions to design-standards, the aesthetic register to creative-direction, the site-structure modeling to information-architecture, and the component implementation to frontend-component-build. It does not duplicate any of them; it composes their outputs into a vertical-correct site.
It is the build-side counterpart to competitor-experience-audit. The audit measures the field on seven dimensions; this skill builds to those same seven, dimension by dimension. The two are deliberately structured as two sides of one standard so an experience bar produced by the audit maps one-to-one onto what this skill builds.
When to use
- Building a new site in a vertical you have not built in before
- Fixing a build that reads as generic or off-vertical even though it is technically clean (the SEO, accessibility, and component implementation are correct, but the page does not read like its category)
- Building to an experience bar produced by
competitor-experience-audit(the audit-fed path is the intended use) - Adding the merchandising, density, and navigation conventions a site in this category is expected to carry
- Building a showcase or demo where the build must read as its vertical, not as a generic SaaS landing page
When NOT to use
- Day-to-day visual decisions on a known build (use
design-standards) - Defining the brand register or aesthetic from scratch (use
creative-direction,brand-archetype-system,brand-discovery) - Building one component well (use
frontend-component-build) - Modeling site structure and navigation hierarchy from scratch (use
information-architecture) - Building a formal design system or token system (use
design-system) - Auditing the competitive field for what conventions exist (use
competitor-experience-audit; this skill consumes that audit's output, it does not produce it)
Required inputs
- The site shape the build is for, from the controlled list:
ecommerce-catalog,inventory-listing,directory-marketplace,local-service-booking,subscription-app,hospitality-food,institution-mission,b2b-manufacturer,ecommerce-standout,hospitality-experience. - An experience bar from
competitor-experience-audit(the cross-site patterns and gaps for the vertical). If absent, the skill falls back to the per-shape default conventions in the references; the audit-fed path is stronger and is the intended use. - The brand register and aesthetic decisions (from
creative-directionor already settled). This skill does not redefine them. - The framework or technical context the build will use (the skill is stack-agnostic but the consumer needs to know).
If the site shape is unclear, ask. Building to the wrong shape's conventions produces a build that reads off-vertical, which is the exact failure this skill exists to prevent.
The framework: build to the seven dimensions
The framework mirrors the seven dimensions competitor-experience-audit measures. Each is expressed here as build guidance (what to do) rather than audit questions (what to observe). Run them in order; the synthesis in dimension 7 is the composition checklist that catches what the earlier six missed.
1. Build the primary task as the dominant, earliest element
The visitor arrives with a primary job. The leaders in the vertical lead with that job; off-vertical builds bury it behind marketing copy or competing CTAs.
- Identify the vertical's primary task from the experience bar or the per-shape reference. (Ecommerce-catalog: a fitment, search, or category-entry interaction. Hospitality-food: a menu or reservation entry. Inventory-listing: a search-and-filter. Local-service-booking: a "book now" or service-picker.)
- Give that task the largest, earliest interactive treatment in the hero.
- Demote competing CTAs. The hero gets one primary action; secondary actions live below the fold or in the chrome.
- Reach the core task without scrolling on the device class the vertical's visitors use.
- For commercial pages, sticky a compact version of the primary task to the chrome so it stays reachable on scroll.
If the primary task cannot be the hero (a constraint from a stakeholder, a regulated vertical), surface the closest equivalent and document the compromise. Do not silently push the primary task below the fold.
2. Build to the vertical's layout register and density
Most off-vertical builds default to a generic SaaS-airy landing page. Retail and catalog verticals are denser; editorial verticals are calmer; hospitality is image-led. Build to the field's density, not to a default.
- Identify the register from the experience bar or the per-shape reference (storefront-dense, editorial-calm, marketing-airy, image-led).
- Match the field's content-per-viewport: a catalog vertical surfaces more modules above the fold than a SaaS landing page.
- Choose the module count above the fold deliberately. State the count in the build notes.
- Route the actual color, type, and aesthetic choice to
creative-directionanddesign-standards. This skill owns the density and module-count decision, not the visual register itself.
Composition decision lives here. Visual register lives there. Do not redefine either; route to the right skill.
3. Build the merchandising and category surface the vertical uses
How much of the catalog or offering is surfaced at once, and in what shape, is a per-vertical decision the build must own.
- Surface the catalog the way the vertical does: how many category, collection, or entry-point modules appear before scroll.
- Signal depth the way the vertical signals it: mega-menu hover, grid of category cards, faceted-search side rail, breadcrumb-driven hierarchy. Pick one and commit; do not stack three.
- Choose flat (one tier visible) or hierarchical (parent plus child categories shown together) per the vertical's convention.
- Use icon / image / text / hybrid category presentation per the vertical.
Cross-reference information-architecture for the structural modeling of the nav tree. This skill owns the compositional surface; that skill owns the structural model.
4. Build both the know-what-I-want and the browse path the vertical expects
Verticals differ on which path is primary. A catalog vertical needs both search and category browse, with search often primary. A hospitality-food site rarely needs search but always needs a menu and a reservation entry. Build to the paths the field actually offers, in the prominence the field gives them.
- List the paths the experience bar names (by category, by brand, by attribute, by vehicle, by location, by occasion, by audience).
- Score each path as primary, secondary, or absent per the vertical.
- Build the primary paths into the hero or primary nav. Build the secondary paths into the chrome or below the fold. Drop the paths the field does not use.
- Search input prominence is a per-vertical decision: header-primary, secondary in a side rail, or absent. Match the field.
Route the structural nav modeling to information-architecture; route the search UX patterns to that skill or to the project's existing search component. This skill owns which paths exist and how prominent each is.
5. Build the brand posture consistently across the page
The brand register is set by creative-direction and brand-archetype-system. This skill does not redefine it. But composition decides whether the posture holds across the page or only in the hero.
- Carry the brand register from the hero into the nav, the cards, the empty states, the trust signals, and the footer. A premium-utilitarian register that disappears below the hero is a composition failure, not a brand failure.
- Apply the voice consistently in headings, microcopy, CTAs, and empty states. Do not let the empty states default to generic copy.
- Hold the imagery posture across the page: a vertical that uses studio product photography in the hero does not switch to abstract illustration in the trust section.
Route the actual register choice and the voice attributes to creative-direction and brand-voice. Own the consistency requirement here.
6. Build the trust and conversion signals the vertical requires
Each vertical has a different set of trust signals it is expected to carry. A build that ships without them reads as not-yet-credible. Build them in.
- Identify the required signals from the experience bar or the per-shape reference. (Ecommerce-catalog: reviews, guarantees, pricing transparency, stock signals, fitment confirmation, shipping promises. Hospitality-food: hours, location, reservation availability, menu price range. B2b-manufacturer: certifications, customer logos, spec sheets. Local-service-booking: licensing, insurance, service-area, response time.)
- Surface them where the field surfaces them. Reviews near the buying action; service-area in the hero; certifications in the chrome.
- Honor the functional-vs-demo line for showcase builds: real where touchable (real reviews if they exist, real cart state), honestly demo where it needs a backend (clearly labeled demo checkout modal, no fake payment).
7. Run the synthesis as a composition checklist
The seventh dimension is the synthesis: the specific conventions a credible build in this vertical must have, named explicitly. Run this as a pre-ship checklist before declaring the composition done.
- Pull the conventions from the experience bar's dimension 7, or from the per-shape reference's recurring-conventions section.
- For each, confirm the build does it. Mark each item present, absent, or routed elsewhere (with a link to the routing decision).
- If three or more conventions are absent, the build is off-vertical. Go back to the earlier dimensions; this is not a polish-pass fix.
- If one or two are absent, name the reason: deliberate positioning (the wedge the build is exploiting from the audit's gaps section), constraint (stakeholder, technical), or oversight (fix before ship).
The synthesis is what catches builds that pass dimensions 1 through 6 individually but compose into something off-vertical anyway. Run it.
How the skill consumes an experience bar
When competitor-experience-audit has run for the vertical, its output is the input to this skill. The audit's seven dimensions map directly onto this skill's seven, in order. Concretely:
- The audit's
Primary-task prominencepattern names what to build in this skill's dimension 1. - The audit's
Layout register and densitypattern names the density target for dimension 2. - The audit's
Merchandising and category surfacepattern names the surface shape for dimension 3. - The audit's
Primary navigation and search pathspattern names the path inventory for dimension 4. - The audit's
Brand register and convictionpattern names the registercreative-directionshould be working from, which this skill carries across the page in dimension 5. - The audit's
Trust and conversion signalspattern names the required signals for dimension 6. - The audit's
Recurring vertical conventionsare the checklist for dimension 7. - The audit's
Gapssection names the positioning opportunity the build can exploit as its wedge.
If no audit has been run, fall back to the per-shape reference files in references/. The default conventions there are derived from observable cross-site patterns and are honest, but a vertical-specific audit is stronger and is the intended path.
Stays in its lane (load-bearing)
This skill owns one thing: vertical-level page and site composition. It routes everything else, and that routing discipline is what keeps it useful and what keeps the catalog clean.
- Visual tokens, color, type, contrast, spacing scale: route to
design-standards. - Aesthetic register, archetype, voice attributes: route to
creative-directionandbrand-archetype-system. - Site structure, nav hierarchy, URL design: route to
information-architecture. - Component API, states, accessibility implementation: route to
frontend-component-build. - Formal design system, token library, component documentation: route to
design-system. - The audit that produces the experience bar: route to
competitor-experience-audit. - On-page SEO and accessibility audits: route to
seo-onpageandaccessibility-audit.
If a build decision sits inside one of those skills, do not restate the standard here. Link to the skill and let it own its territory.
Workflow
- Confirm the site shape and the experience bar. Site shape from the controlled list. Experience bar from
competitor-experience-auditif run; the per-shape reference otherwise. - Run dimensions 1 through 6 in order. For each, name the build decision and which other skill it routes to (for the parts that route elsewhere).
- Run dimension 7 as a checklist. Confirm every recurring convention from the vertical is present, absent for a stated reason, or routed elsewhere.
- Write the composition notes. Use the template in
references/shape-conventions-checklist.md. The notes are whatfrontend-component-buildanddesign-standardsread as the spec to implement against. - Hand off to the implementation skills. This skill's output is composition decisions, not finished components. The build is done when those decisions have been implemented through
frontend-component-build,design-standards, and the project's component library.
Failure patterns
When you spot one of these, push back before delivering.
- "Make it look modern." Vague. Ask which vertical, which experience bar, which conventions. "Modern" without a vertical produces the generic SaaS-airy build this skill exists to prevent.
- Building to a generic / SaaS-airy default when the vertical wants a denser storefront. Retail and catalog verticals surface more per viewport than a marketing page. If the build looks calm and roomy, check the density against the field.
- Burying the primary task. A hero stuffed with marketing copy and three competing CTAs is the most common off-vertical failure. Dimension 1 is not optional.
- Redefining aesthetic vocabulary instead of routing to
creative-direction. This skill does not coin new register names or invent new voice attributes. If the question is what posture the brand takes, that iscreative-direction's call. - Treating the experience bar as optional when it is available. If
competitor-experience-audithas run for the vertical, the bar is the spec. Building to the per-shape default while ignoring the audit's bar is half a build. - Producing a template instead of building to the field's actual conventions. Per-shape references are starting points, not finished templates. They name the patterns the field shares; the build still has to land them in this brand and this brief.
- Skipping dimension 7. A build that passes dimensions 1 through 6 individually can still compose off-vertical. The synthesis check is the one that catches it.
- Calling implementation detail "composition". Button radius, font size, color contrast are
design-standardsdecisions. Module count, primary-task prominence, category-surface depth are this skill's decisions. Do not bleed across the line.
Output format
Default output is a markdown set of composition notes at composition-[site-shape]-[brand-slug].md in the project root. Structure:
- The site shape and the experience bar consulted (or "no audit available; per-shape reference used")
- Dimension-by-dimension build decisions (1 through 6)
- The synthesis checklist (dimension 7), each convention marked present / absent / routed
- The wedge (which gap from the audit's gaps section this build is exploiting, if any)
- The hand-off list: what
frontend-component-build,design-standards, andcreative-directionneed from these notes to implement
Keep notes under 1500 words. The notes are a spec for the implementation skills; if they need more detail, that is a sign one of those skills should own the deeper specification.
Reference files
references/shape-conventions-checklist.md- The per-dimension build checklist applied as a pre-ship review. Mirror of the audit's dimension checklist on the build side.references/shape-ecommerce-catalog.md- Default conventions for the ecommerce-catalog shape (first test case; auto-parts and similar dense storefronts).references/shape-inventory-listing.md- Stub; fill when the first inventory-listing demo is built.references/shape-directory-marketplace.md- Stub.references/shape-local-service-booking.md- Stub.references/shape-subscription-app.md- Stub.references/shape-hospitality-food.md- Stub.references/shape-institution-mission.md- Stub.references/shape-b2b-manufacturer.md- Stub.references/shape-ecommerce-standout.md- Stub.references/shape-hospitality-experience.md- Stub.
GitHub 저장소
자주 묻는 질문
vertical-site-conventions Skill이란 무엇인가요?
vertical-site-conventions은(는) rampstackco이(가) 만든 Claude Skill입니다. Skill은 Claude가 필요할 때 불러오는 지침과 리소스를 묶어 추가 프롬프트 없이 vertical-site-conventions 관련 작업을 수행할 수 있게 합니다.
vertical-site-conventions은(는) 어떻게 설치하나요?
이 페이지의 설치 명령을 사용하세요. vertical-site-conventions을(를) Claude Code 플러그인으로 추가하거나 저장소를 skills 디렉터리에 복제한 다음 Claude를 다시 시작해 Skill을 불러옵니다.
vertical-site-conventions은(는) 어떤 카테고리에 속하나요?
vertical-site-conventions은(는) 메타 카테고리에 속합니다.
vertical-site-conventions은(는) 무료로 사용할 수 있나요?
네. vertical-site-conventions은(는) AIMCP에 등록되어 있으며 무료로 설치할 수 있습니다.
연관 스킬
이 스킬은 콘텐츠 콜렉션(Content Collections)을 위한 프로덕션 검증된 설정을 제공합니다. 콘텐츠 콜렉션은 Markdown/MDX 파일을 Zod 검증이 포함된 타입 안전한 데이터 콜렉션으로 변환해주는 TypeScript 최우선 도구입니다. 블로그, 문서 사이트 또는 콘텐츠 중심의 Vite + React 애플리케이션을 구축할 때 타입 안전성과 자동 콘텐츠 검증을 보장하기 위해 사용하세요. Vite 플러그인 구성과 MDX 컴파일부터 배포 최적화 및 스키마 검증에 이르기까지 모든 것을 다룹니다.
이 스킬은 개발자들이 Polymarket 예측 시장 플랫폼을 활용한 애플리케이션을 구축할 수 있도록 지원하며, 거래 및 시장 데이터를 위한 API 통합 기능을 포함합니다. 또한 WebSocket을 통한 실시간 데이터 스트리밍을 제공하여 실시간 거래와 시장 활동을 모니터링할 수 있습니다. 이를 통해 거래 전략을 구현하거나 실시간 시장 업데이트를 처리하는 도구를 생성하는 데 활용할 수 있습니다.
이 스킬은 개발자들이 명령어, 파일, LSP 작업 등 25개 이상의 이벤트 유형에 연결되는 OpenCode 플러그인을 만들 수 있도록 돕습니다. JavaScript/TypeScript 모듈을 위한 플러그인 구조, 이벤트 API 명세, 구현 패턴을 제공합니다. OpenCode AI 어시스턴트의 라이프사이클을 사용자 정의 이벤트 기반 로직으로 가로채거나, 모니터링하거나, 확장해야 할 때 사용하세요.
SGLang은 RadixAttention 프리픽스 캐싱을 활용하여 JSON, 정규식, 에이전트 워크플로우를 위한 고속 구조화 생성에 특화된 고성능 LLM 서빙 프레임워크입니다. 특히 반복되는 프리픽스가 있는 작업에서 상당히 빠른 추론 속도를 제공하여 복잡한 구조화 출력 및 다중 턴 대화에 이상적입니다. 제약 디코딩이 필요하거나 광범위한 프리픽스 공유가 있는 애플리케이션을 구축할 때는 vLLM과 같은 대안보다 SGLang을 선택하십시오.
