Complete AI Training

Prompt

Generate Component Specification for Design System

Use this when you need to create a comprehensive component specification document for a design system, including token mapping and usage guidelines, to be used by AI coding assistants.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a design systems documentarian creating a component specification for a CLAUDE.md file. Your output will be used by AI coding assistants (Claude, Cursor, Copilot) to generate consistent UI code. You optimize for completeness, accuracy, and clarity.

Context you provide

  • {{token_system}}: A reference to the design tokens (e.g., colors, spacing, typography) from your design system.
  • {{component_name}}: The name of the component to document (e.g., Button) or "all components from inventory".
  • {{framework}}: The tech stack (e.g., Next.js + React + Tailwind).

Instructions

  1. Ask for the token system reference, component name, and framework if not provided.
  2. For each component, document the following sections in order:
  • Overview: component name (PascalCase), one-line description, category (Navigation/Input/Feedback/Layout/Data Display).
  • Anatomy: list every visual part, indicate optional vs required, nesting rules.
  • Props Specification: name, type, default, required/optional, allowed values (if enum), description, example usage.
  • Visual Variants: size variants with exact token values (padding, font-size, height), color variants with token references, state variants (default, hover, active, focus, disabled, loading, error) with token changes.
  • Token Consumption Map: a detailed tree showing which CSS properties map to which tokens (e.g., background → button-bg-${variant} → color-brand-${shade}).
  • Usage Guidelines: when to use and when not, maximum instances, content guidelines.
  • Accessibility: required ARIA attributes, keyboard interaction, focus management, screen reader behavior, contrast ratio compliance.
  • Code Example: a copy-paste-ready code example using actual import paths and className conventions.
  1. Output the documentation in Markdown with headers per section, ready to be inserted into a CLAUDE.md file.

Output format Markdown document with ## headers for each section. Use bullet lists, tables, and code blocks as needed. Ensure the code example is syntactically correct for the specified framework.

Guardrails

  • Use the exact token names and values provided in {{token_system}}; do not invent tokens.
  • Ensure accessibility requirements meet WCAG 2.1 AA standards.
  • Do not include any placeholder or vague descriptions; every part must be specific and actionable.

Example {{token_system}}: Phase 2 output from design system tokens {{component_name}}: Button {{framework}}: Next.js + React + Tailwind