← Back to Skills Marketplace
ivangdavila

Design System

by Iván · GitHub ↗ · v1.0.0
linuxdarwinwin32 ✓ Security Clean
853
Downloads
0
Stars
12
Active Installs
1
Versions
Install in OpenClaw
/install design-system
Description
Build design systems with tokens, components, and documentation that scale across teams and products.
README (SKILL.md)

Setup

On first use, read setup.md for integration guidelines. All preferences are stored in ~/design-system/memory.md.

When to Use

User needs to create, maintain, or extend a design system. Agent handles token architecture, component patterns, documentation structure, and cross-platform consistency.

Architecture

Memory lives in ~/design-system/. See memory-template.md for structure.

~/design-system/
├── memory.md         # Status + context + decisions
└── tokens/           # Token definitions if exported

Quick Reference

Topic File
Setup process setup.md
Memory template memory-template.md

Core Rules

1. Tokens First, Components Second

Design tokens are the foundation. Before building any component:

  • Define color tokens (semantic, not raw hex)
  • Define spacing scale (consistent multiplier)
  • Define typography scale (modular)

Components consume tokens. Never hardcode values.

2. Semantic Over Literal Naming

Bad Good
blue-500 primary
14px text-sm
8px space-2

Semantic names survive rebrand. Literal names break everything.

3. Three-Tier Token Architecture

Primitive → Semantic → Component
   ↓           ↓          ↓
 gray-900   text-primary  button-text
  • Primitive: Raw values (colors, sizes)
  • Semantic: Meaning-based (primary, danger, muted)
  • Component: Specific use (button-bg, card-border)

4. Document Decisions, Not Just Specs

Every token and component needs:

  • What: The value or pattern
  • When: Usage context
  • Why: The decision behind it
  • When NOT: Anti-patterns to avoid

5. Platform-Agnostic Source of Truth

Design tokens should export to:

  • CSS custom properties
  • Tailwind config
  • iOS/Android native
  • Figma variables

One source, many outputs. Use Style Dictionary or similar.

6. Component API Consistency

All components follow the same patterns:

  • Same prop naming (variant, size, disabled)
  • Same size scale (sm, md, lg)
  • Same variant names (primary, secondary, ghost)

Predictability beats cleverness.

7. Versioning and Migration

Breaking changes need:

  • Version bump (semver)
  • Migration guide
  • Deprecation warnings before removal
  • Codemods when possible

Common Traps

  • Premature abstraction → Build 3 instances before extracting a pattern
  • Token explosion → 50 grays is not a system, it is chaos
  • Skipping documentation → Undocumented patterns get reimplemented wrong
  • Designing for edge cases first → Cover 80% well before 100% poorly
  • No dark mode strategy → Retrofit is 10x harder than planning upfront
  • Inconsistent spacing → Use a scale (4px base), not arbitrary values
  • Component prop sprawl → More than 10 props means split the component

Security & Privacy

Data that stays local:

  • Design decisions in ~/design-system/
  • Token definitions and component specs

This skill does NOT:

  • Access files outside ~/design-system/
  • Make network requests
  • Store sensitive data

Related Skills

Install with clawhub install \x3Cslug> if user confirms:

  • css — Styling fundamentals
  • tailwindcss — Utility-first CSS
  • frontend — Frontend development
  • ui — User interface patterns
  • design — Design principles

Feedback

  • If useful: clawhub star design-system
  • Stay updated: clawhub sync
Usage Guidance
This skill appears coherent and low-risk: it only uses local files and includes templates for storing design-system memory at ~/design-system/memory.md. Before installing or using it, consider: 1) inspect the contents of ~/design-system/ if you want to confirm what the agent wrote; 2) do not paste secrets or credentials into the memory files (they are persistent); 3) the skill claims it won’t make network requests, but the platform hosting the agent may allow network access — if you need to enforce no-network behavior, verify platform/network policies; 4) if you don’t want any persistent files, avoid enabling the skill or delete ~/design-system/ after use. If the skill later requests environment variables, executables, or external URLs, treat that as a red flag and re-evaluate.
Capability Analysis
Type: OpenClaw Skill Name: design-system Version: 1.0.0 The skill bundle is benign. All files (SKILL.md, memory-template.md, setup.md) consistently describe an AI agent's role in assisting with design system creation, focusing on local memory management within `~/design-system/` and user interaction. Crucially, SKILL.md explicitly states under 'Security & Privacy' that the skill 'does NOT: Access files outside ~/design-system/, Make network requests, Store sensitive data', which acts as a clear instruction to the agent against malicious behaviors and aligns with a benign purpose. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection attempts to subvert the agent, or obfuscation.
Capability Assessment
Purpose & Capability
Name/description (design system tokens, components, docs) align with the instructions and included templates. No unrelated binaries, env vars, or services are requested.
Instruction Scope
SKILL.md and setup.md limit behavior to asking the user questions, advising on patterns, and reading/writing files in ~/design-system/. The skill explicitly states it will not access files outside that directory or make network requests.
Install Mechanism
No install spec and no code files — instruction-only means nothing is downloaded or written by an installer. Lowest-risk install posture.
Credentials
No environment variables, credentials, or config paths are requested. All required state is local and file-based (~/design-system/). Proportional to the claimed functionality.
Persistence & Privilege
The skill persists a memory file under the user's home (~ /design-system/memory.md). This is reasonable for a design-system assistant, but it is persistent data stored on disk — users should be aware of what is written and avoid saving secrets there.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install design-system
  3. After installation, invoke the skill by name or use /design-system
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug design-system
Version 1.0.0
License
All-time Installs 12
Active Installs 12
Total Versions 1
Frequently Asked Questions

What is Design System?

Build design systems with tokens, components, and documentation that scale across teams and products. It is an AI Agent Skill for Claude Code / OpenClaw, with 853 downloads so far.

How do I install Design System?

Run "/install design-system" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Design System free?

Yes, Design System is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Design System support?

Design System is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).

Who created Design System?

It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.

💬 Comments