Analysis updated 2026-06-24
Drop a brand DESIGN.md into a project so an AI agent builds UI in that style
Clone the visual feel of Stripe, Figma, or Airbnb without writing a design brief
Keep AI generated colors, typography, and spacing consistent across a product
Study public design systems in a structured markdown format
| khalidabdi1/design-ai | limin112/wechat-publish-template | orchestration-agent/agentorchestration | |
|---|---|---|---|
| Stars | 152 | 149 | 155 |
| Language | — | HTML | Python |
| Setup difficulty | easy | easy | hard |
| Complexity | 1/5 | 1/5 | 4/5 |
| Audience | vibe coder | writer | ops devops |
Figures from each repo's GitHub metadata at analysis time.
No install step, just copy a DESIGN.md file into your project and point your AI agent at it.
Design AI Systems is a collection of DESIGN.md files that describe the visual style of well known websites and brands so that AI coding tools can build user interfaces that look like them. The idea is simple: instead of telling an AI agent to make something that feels like Stripe or Airbnb in vague words, you drop in a markdown file that spells out the colors, fonts, spacing, and component patterns of that brand, and the agent has a real reference to work from. A DESIGN.md is a plain text design system document. The README explains that it sits next to AGENTS.md, which tells the agent how to work in a repo, while DESIGN.md tells the agent how the interface should look and feel. The format is human readable, version controlled with Git, easy to edit by hand, and meant to be portable across projects. Every file in the library follows a fixed nine section structure. The sections cover visual theme and atmosphere, color palette and roles, typography rules, component stylings, layout principles, depth and elevation, do's and don'ts, responsive behavior, and a final agent prompt guide with ready to use snippets. The repository ships with around 150 of these files, grouped by category such as social, creative platforms, commerce, media and gaming, and developer tools. Featured examples include Stripe, Figma, Apple, Slack, Netflix, Airbnb, Coinbase, PayPal, Asana, and Zoom. To use one, you pick a company folder, copy its DESIGN.md into your own project root, and tell your AI agent to treat that file as the design source of truth.
Library of about 150 DESIGN.md files describing the colors, typography, and components of well known brands so AI coding agents can generate UI matching those styles.
MIT license, free to use, modify, and redistribute including for commercial work, with attribution.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.