What is the design-system-generator skill?
Design-system-generator is a skill in the awesome-ai-agent-skills collection. Its job is a durable design system document grounded in your real stack. The agent reads existing docs, screens, components and tokens first, then writes the project contract, and verifies paths, token names and rendered output before reporting.
Design-system-generator ships inside the project-development-skills bundle of thienanblog's awesome-ai-agent-skills repository. It is for one situation: the durable design system document is the deliverable, not a one-off UI edit.
The description spells out the scope: tokens, components, accessibility, motion and visual verification. The agent is told to keep the output concise and tied to the product's actual stack and visual language.
What the agent does, in order
- Inspect first. It reads existing design docs, brand references, representative screens, shared components, theme configuration, tokens and build tooling, and infers framework, CSS approach, themes and export formats.
- Preserve what works. An established, coherent system is kept. For a new product, it recommends one direction within your constraints.
- Write the contract, using a bundled template as a menu of topics rather than a form to fill.
- Verify paths, token names, component APIs and build claims against their owners, then report.
What the document covers
- Surfaces, themes, devices, and any language or right-to-left constraints.
- Token ownership, typography, color, spacing, layout and motion.
- Shared components and the interaction states each one really supports.
- Extension rules for tokens, components, utilities and local exceptions.
- Keyboard interaction, visible focus, contrast targets, semantic markup and reduced motion.
- Visual verification at comparable states and viewports, plus asset loading and cache invalidation when the design system owns them.
Where the skill stops
- It is not for a single screen. A one-off UI edit does not need a whole document.
- It does not add a component library or migrate build tooling just to match its own examples.
- It omits prompt-writing advice and generic framework tutorials. For component rules see shadcn, and for a Figma-side library see figma-generate-library.
Use it with UiChemy
A written design system gives your agent fixed tokens to build against. UiChemy then moves the finished build into WordPress: AI to WordPress imports an AI build, and Figma to WordPress converts a Figma design. Click Get Started on uichemy.com. For the colors, see color-mode-and-theme.
How to install Design System Generator Skill: Write a DESIGN_SYSTEM.md
/plugin marketplace add thienanblog/awesome-ai-agent-skillsTo refresh later, the README gives /plugin marketplace update for Claude Code and codex plugin marketplace upgrade awesome-ai-agent-skills for Codex.Add the marketplace
In Claude Code, run
/plugin marketplace add thienanblog/awesome-ai-agent-skills. The README also givesclaude plugin marketplace add thienanblog/awesome-ai-agent-skillsfor the terminal, which the desktop app needs before its plugin browser lists the marketplace.Install the plugin that holds the skill
Run
/plugin install project-development-skills@awesome-ai-agent-skills. The design-system-generator skill is one of the skills bundled in project-development-skills.Or use Codex
Run
codex plugin marketplace add thienanblog/awesome-ai-agent-skills, open/pluginsinside codex, choose the Awesome AI Agent Skills marketplace and select Install plugin on project-development-skills.Ask for the document
Describe the deliverable, for example a design system document for this project, with a pointer to your tokens and components. Specialist skills are discovered from their descriptions.
Check that it loaded
Ask for a DESIGN_SYSTEM.md on a small project. The agent should first read your tokens, components and theme config, then write the document and report what it verified.
Key features
- Inspect before proposingThe agent reads design docs, brand references, representative screens, shared components, theme config, tokens and build tooling before it writes, and asks only about unresolved product choices that matter.
- Preserves an established systemA coherent existing system is kept as is. For a new product, the skill recommends one suitable direction and compares alternatives only when useful or requested.
- Template as a menuTEMPLATE_DESIGN_SYSTEM.md lists topics to choose from. Inapplicable sections are dropped and examples are replaced with verified project decisions.
- Accessibility baked inThe contract covers keyboard interaction, visible focus, contrast targets, semantic markup and reduced motion, not as an afterthought.
- Project specific, not genericStatements must be actionable and specific to the product. The document serves the product team, so framework tutorials and unused frameworks or themes are left out.
- Token exports on requestCSS or JSON token files are produced only when requested or needed. Example files tokens.css and design-tokens.json are read only for those formats.
- Discoverability patchIf an instruction-file link helps agents find the document, a minimal pattern in examples/AGENTS.patch.md is used, preserving any existing instruction file ownership or symlink.
- Build-aware, not build-forcingHashed filenames and manifests are treated as one cache strategy, not a required new build system. Pipeline detail links to your existing build guidance.
- Verification stepThe agent checks documented paths, token names, component APIs and build claims, inspects rendered components when visual decisions are made, and reports when visual validation could not run.
- No subagents without approvalIt runs in the main conversation by default. Delegation can raise usage, so subagents need explicit approval for the agent count and scope.
Frequently asked questions
What is the design-system-generator skill?
It is a skill in thienanblog's awesome-ai-agent-skills collection. It makes an AI coding agent inspect your project and write or revise a concise design system document covering tokens, components, accessibility, motion and visual verification, based on what your product actually uses.
How do I install the design-system-generator skill?
Add the marketplace with /plugin marketplace add thienanblog/awesome-ai-agent-skills, then install project-development-skills@awesome-ai-agent-skills. Codex users add the same repository with codex plugin marketplace add and install the plugin from the /plugins browser. The skill ships inside that bundle.
Which AI coding agents can use the design-system-generator skill?
The README says the collection works with Claude Code, Cursor, Kilo Code, Windsurf, OpenAI Codex and any tool that supports skills or custom instructions. The repo also ships a Codex marketplace and an openai.yaml file inside this skill's folder.
Where does the skill save the design system?
It updates the existing design system owner in your repository, or creates DESIGN_SYSTEM.md at the repository's conventional location. CSS or JSON token exports are written only when requested or needed by the agreed implementation.
Does it work for an existing project that already has a design system?
Yes. The skill is told to preserve an established, coherent system, and to inspect docs, screens, components and tokens first. It revises the document rather than replacing a system that already works. For a related MCP, see website-design-systems-mcp.
Does the design-system-generator skill use subagents?
Not by default. It runs in the main conversation, because delegation can increase usage. It needs explicit approval for the proposed agent count and scope before using subagents, and asks again before either grows.
What license does the design-system-generator skill use?
The awesome-ai-agent-skills repository is licensed Apache-2.0, per the GitHub license field. Check the repository again before you redistribute the skill folder inside a product, since the plugin packages are generated copies of the canonical skill.


