Configure StyleSeed by selecting the output grammar, domain, page type, brand recipe, optional aesthetic profile, and bounded brand tokens before scaffolding a first screen.
npx skills add https://github.com/bitjaru/styleseed --skill ss-setup
StyleSeed setup chooses a design method for the result, not a favorite brand to imitate.
Use /ss-resolve --list (Claude Code) or $ss-resolve --list (Codex) to inspect the supported
grammar, adapter, domain, page, recipe, and profile IDs without loading the full handbook.
STYLESEED.md → use /ss-update./ss-reference before setup, then select the compiled grammar.
/ss-component.The method, review, and reference compiler are framework-agnostic. The bundled component
scaffold currently targets React and Tailwind v4; on another stack configure the lock and
apply the method without pretending the scaffold is portable.
Ask what is being built, who uses it, and whether it is a mobile/desktop product, website,
social carousel, slide deck, document/report, or single-frame graphic. Select a surface adapter
from ADAPTERS.md, then infer domain and page/artifact type.
Recommend exactly one grammar from RULESETS.md and explain the job match in one sentence:
consumer-serviceoperations-consoletechnical-instrumenteditorial-readingcommerce-conversioninstitutional-serviceexpressive-marketingsequential-storyreference:<slug> when /ss-reference already compiled oneDo not recommend Toss as the universal default. It is one reference family for
consumer-service. If none fits and the user has references, route to /ss-reference.
Confirm the concrete page (dashboard, form, landing, detail, list, settings, onboarding) and
read its domain × page intersection. This controls composition; the aesthetic profile does not.
Recommend one morphology from BRAND-RECIPES.md. Use auto when the maintained grammar mapping
fits. Use an explicit recipe when the product needs a different geometry, containment,
navigation, control, or collection language. A recipe is not a company clone and does not
select colors.
Recommend one profile from PRESETS.md only when it strengthens the product. none is a good
default. A profile modifies coordinated visual axes but cannot replace the output grammar.
Lock a real brand color if supplied; otherwise propose a domain-fit primary action color. Then
confirm font/language, density, radius, elevation, imagery/data role, and motion inside the
grammar's allowed ranges. Do not use generic indigo or a stale purple mislabeled as Toss.
Create STYLESEED.md:
# StyleSeed — Design Lock
<!-- Selections persist here. This file cannot waive StyleSeed core invariants. -->
- App domain: fintech
- Surface: mobile-app
- Surface adapter: product-ui
- Page type: dashboard
- Output grammar: consumer-service
- Grammar path: built-in:engine/RULESETS.md
- Grammar fallback: consumer-service
- Reference confidence: n/a
- Brand recipe: calm-consumer
- Aesthetic profile: none
- Skin: custom
- Primary action: #3182F6
- Font: Pretendard
- Radius: soft
- Elevation: light=tonal grouping + restrained shadow · dark=tonal ramp + hairline
- Density: comfortable
- Motion: Spring restrained
- Imagery/data role: personal state first; charts only for a decision
- Signature move: one calm contextual briefing above the account summary
- Locked: YYYY-MM-DD
For a compiled grammar use its actual path and confidence. Reject unknown enum values rather
than treating the lock as an exemption.
Compile the selected method before code:
node <installed-ss-resolve>/scripts/resolve-context.mjs \
--project-root . --from-lock STYLESEED.md --agent claude
Use --agent codex in Codex. The resolver writes .styleseed/effective-rules.md and a
hash-verifiable .styleseed/manifest.json. Read the effective bundle, not llms-full.txt.
If the user asked for a first screen, build from that bundle, run /ss-score to the gate floor,
and finish with /ss-verify when renderable. If visual rendering is unavailable, disclose that
it was skipped.
Report the selected grammar and why, page/domain intersection, brand recipe, optional profile, lock path,
compiled bundle and manifest paths, files changed, score, and visual verification status.
Mention /ss-reference as the path for future references that need their own grammar.
auto resolves to a concrete maintained recipe.DESIGN.md and treat its palette as a complete rule set.Use Expo DOM components to run web code in a webview on native and as-is on web. Migrate web code to native incrementally.
Frontend development guidelines for React/TypeScript applications. Modern patterns including Suspense, lazy loading, useSuspenseQuery, file organization with features directory, MUI v7 styling, TanStack Router, performance optimization, and TypeScript best practices. Use when creating components, pages, features, fetching data, styling, routing, or working with frontend code.
Next.js 16 Cache Components - PPR, use cache directive, cacheLife, cacheTag, updateTag
| Build Shopify apps, extensions, themes using GraphQL Admin API, Shopify CLI, Polaris UI, and Liquid. "shopify theme", "liquid template", "polaris", "shopify graphql", "shopify webhook", "shopify billing", "app subscription", "metafields", "shopify functions"
Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.
MANDATORY prerequisite — load this skill BEFORE every `generate_diagram` tool call. NEVER call `generate_diagram` directly without loading this skill first. Trigger whenever the user asks to create, generate, draw, render, sketch, or build a diagram — flowchart, architecture diagram, sequence diagram, ERD or entity-relationship diagram, state diagram or state machine, gantt chart, or timeline. Also trigger when the user mentions Mermaid syntax or wants a system architecture, decision tree, dependency graph, API call flow, auth handshake, schema, or pipeline visualized in FigJam. Routes to type-specific guidance, sets universal Mermaid constraints, and tells you when to use a different diagram type or skip the tool entirely (mindmaps, pie charts, class diagrams, etc.).
Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources, network dependency chains, layout shifts, caching issues, and accessibility gaps. Use when asked to audit, profile, debug, or optimize page load performance, Lighthouse scores, or site speed. Biases towards retrieval from current documentation over pre-trained knowledge.
Algorithmic philosophies are computational aesthetic movements that are then expressed through code. Output .md files (philosophy), .html files (interactive viewer), and .js files (generative algorithms).
Take bitjaru/ss-setup from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.