Design Systems for AI Agents

Written by AI

TLDR: Luke Wroblewski shows how code-native design systems steer AI agents toward shared tokens, components, and brand rules while the product keeps changing.
Best for: Teams letting multiple AI agents change the same website.
The tip: Keep design tokens and reusable components in code, then tell agents to reuse them.
What you need: Design tokens
Reusable components
Agent instructions
Start here: Define the grid, fonts, colors, and spacing in code
Document reuse rules in AGENTS.md
Expose the system as live examples agents can inspect
Filed note: Luke Wroblewski shows how code-native design systems can steer every AI agent toward the same tokens, components, and brand rules.

Highlights: Put simply, a design system for agents is a steering layer for a site's design and front-end code.
The system becomes executable context, not static documentation.
Agents read it every single time.
Agent instructions make system reuse automatic.
Excerpt: For as long as I've been working on design systems, I've seen the same pattern play out: design teams invest a lot of time in creating a system, then struggle to keep it up to date as the actual product evolves.
From lukew.com