惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

G
Google Developers Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
WordPress大学
WordPress大学
阮一峰的网络日志
阮一峰的网络日志
V
Visual Studio Blog
雷峰网
雷峰网
博客园_首页
The Cloudflare Blog
Hugging Face - Blog
Hugging Face - Blog
酷 壳 – CoolShell
酷 壳 – CoolShell
爱范儿
爱范儿
小众软件
小众软件
D
Docker
P
Proofpoint News Feed
B
Blog
Vercel News
Vercel News
B
Blog RSS Feed
U
Unit 42
月光博客
月光博客
The GitHub Blog
The GitHub Blog
Apple Machine Learning Research
Apple Machine Learning Research
Y
Y Combinator Blog
I
InfoQ
Recent Announcements
Recent Announcements

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
What Is DESIGN.md? A Practical Guide to Design Systems fo...
PromptMaster · 2026-06-28 · via DEV Community
Cover image for What Is DESIGN.md? A Practical Guide to Design Systems for AI Agents

PromptMaster

DESIGN.md is an open format from Google Labs that describes a design system to AI coding agents. It pairs machine-readable design tokens (YAML front matter) with human-readable rationale (markdown prose), so an agent generates UI that matches your brand instead of generic defaults.

Why agents need it

Left alone, an AI coding agent draws on the most common patterns in its training data — generic blues, default fonts, arbitrary spacing. It has no way to know your design system unless you tell it, every single time. The result is UI that looks plausible but not like yours.

DESIGN.md removes the "every single time." You describe your visual identity once, in a file the agent reads before generating any UI.

Where it lives

DESIGN.md is a plain file in your repo root. Because it is markdown with YAML front matter — formats every developer knows — it needs no build step and no special viewer. The agent reads it as-is, and it travels with your project.

The two parts

---
name: Heritage
colors:
  primary: "#1A1C1E"
  tertiary: "#B8422E"
typography:
  h1:
    fontFamily: Public Sans
    fontSize: 3rem
---

## Colors
High-contrast neutrals with a single warm accent. Use the
accent (tertiary) only for the primary action - never decoratively.

The front matter gives the agent exact values. The prose gives it the rules. Together they let the agent apply your system correctly, not just access its raw numbers.

How it relates to CLAUDE.md and AGENTS.md

If you use CLAUDE.md or AGENTS.md, this will feel familiar. DESIGN.md is to design what those files are to code conventions — persistent context the agent reads on every interaction. Complementary layers that do not overlap.

FAQ

Is it free? Yes — open format, free CLI.
Do I need a build pipeline? No. The agent reads the file as-is.
Which agents work with it? Any with a persistent-context mechanism: Claude Code, Cursor, Kiro, Windsurf.

The bottom line

DESIGN.md is a small, durable investment: describe your design system once, and every UI an agent generates starts from your brand instead of a generic default.


Free starter: The format, a complete annotated example, and the core idea are on a free cheat sheet: DESIGN.md Quick-Start Cheat Sheet

Go deeper: The full guide covers the entire format — the token schema, the CLI in depth, accessibility, Tailwind and DTCG export, agent integration, and a complete walkthrough: DESIGN.md: The Complete Guide to Design Systems for AI Agents

Have you tried DESIGN.md yet, or are you still re-explaining your design system to your agent every session? Curious where people are in the comments.