An open, community-driven standard for AI-readable, cross-framework component documentation.
The Open Component Documentation Standard (OCDS) defines a structured, machine-readable format for documenting design system components.
As AI becomes embedded in how we design and build digital products, component documentation must evolve beyond human-readable guidelines. AI systems — and future tooling — require explicit, structured metadata to generate accessible, design-consistent interfaces without guesswork.
OCDS provides that structure.
Most design systems already document:
- Component APIs (props, inputs, variants)
- States and behaviors
- Design tokens
- Accessibility requirements
- Usage and composition rules
However, this information is:
- Written primarily for humans
- Inconsistent across organizations
- Difficult for AI tools to interpret
- Framework-specific and non-portable
We believe these shared patterns can be translated into a common, open, extensible standard.
If component documentation becomes structured and machine-readable:
- AI-assisted UI generation becomes reliable
- Design-to-code alignment improves
- Accessibility expectations become enforceable
- Teams can move across companies and systems with less friction
- Interoperability across frameworks becomes achievable
OCDS v0.1 introduces a framework-agnostic core model for:
- Props / inputs
- Variants
- States
- Events
- Slots and parts
- Token references
- Allowed and forbidden combinations
- Conditional constraints
- Accessibility requirements
- Behavioral expectations
- Semantic role definitions
- Name computation guidance
- Keyboard interaction expectations
- Focus management metadata
Mappings for:
- React
- Vue
- Angular
- Flutter
- iOS / Android
- Other platforms