Skip to content

Repository files navigation

Sora Lattice

Sora Lattice generates design-system foundations from a few brand choices — primary/secondary/accent colors, neutrals, status hues, fonts, density and surface style — and emits semantic tokens tuned for WCAG AA contrast. It ships as a Bun workspace orchestrated by Turborepo, split into a reusable token-generation library and an Astro web app that wraps it in a marketing site, an interactive Configurator, live previews, and one-click exports.

Repository layout

Path Purpose
packages/generator @soralabsoss/generator — pure TypeScript library that produces OKLCH ramps, semantic token sets, and CSS / DTCG / Tailwind / shadcn exports. No DOM dependencies; usable from Node or the browser.
apps/web Astro 5 + React 19 frontend: marketing pages, the /generate Configurator, live component playground, and export dialog. Consumes @soralabsoss/generator via workspace:*.

Requirements

  • Node.js 18.20.8+ or 22+ (22+ recommended; pinned in apps/web/.nvmrc).
  • Bun 1.3+ (packageManager in root package.json pins bun@1.3.5).

Getting started

bun install
bun run dev      # starts the web app via Turborepo (Astro on http://localhost:4321)

Other root scripts (all delegate through Turborepo):

bun run build    # build the generator package, then the web app
bun run test     # run vitest in every workspace that defines `test`
bun run preview  # serve the production web build

Turbo task wiring lives in turbo.json; ^build ensures @soralabsoss/generator is compiled before apps/web consumes it.

Working on a single package

bun run --filter @soralabsoss/generator build
bun run --filter @soralabsoss/generator test
bun run --filter @sora-lattice/web dev

See each package's README for API docs, scripts, and customization notes.

About

OKLCH-based design-token generator with WCAG-AA validated theming — turn one brand color into a full CSS, Tailwind, and shadcn token system.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages