ai-native-design-framework
Health Gecti
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Community trust — 21 GitHub stars
Code Gecti
- Code scan — Scanned 7 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
Bu listing icin henuz AI raporu yok.
Make generic UI look consistent. No hardcoding, no ignoring your design system. Spec for AI-native design systems: tokens, slots and rules agents can discover and validate via MCP (coming in v2).
AI-Native Design Framework (AINDF)
Make generic UI look consistent. No hardcoding, no ignoring your design system.
AINDF is a spec, not a component library. It describes how to write down your
design system (tokens, components, slots and the rules between them) as JSON
files that an AI agent can read and check its work against. You keep your own
tokens, components and brand.
Website: https://aindf.oleg.design
What's in this repo
SPEC.md: the spec itself.schemas/: JSON Schemas for the files your design system
publishes: tokens, taxonomy, slots, applicability, presets and patterns.patterns/starter.json: an example pattern for a
newsletter signup section.
The validator, MCP server and builder come in the next release, see
Coming in v2. For now you can check your files against the
schemas.
Getting started
Read
SPEC.md. Sections 3 to 6 cover component layers, tokens,
modifiers, presets and patterns.Describe your design system in JSON files that follow the schemas.
Check each file with any JSON Schema validator that supports draft 2020-12.
With Node installed, this checks the example pattern:npx ajv-cli@5 validate --spec=draft2020 -s schemas/patterns.schema.json -d patterns/starter.jsonSwap in your own schema and data file.
Coming in v2
The next release adds three tools that work together:
- MCP server. Your AI agent looks up what your design system offers
(components, slots, variants, tokens) through MCP instead of guessing. - Validator. Checks a screen config before it is built. It rejects unknown
components, slots or modifiers, one-off styles and code inside the config. - Builder. Builds the screen from that config and a specific version of your
design system. The agent edits the config, not HTML, CSS or JSX.
If your design system is missing something, the agent asks for it to be added
instead of writing its own CSS.
Why it matters: fix something once in the design system and every screen that
uses it should get the fix.
To hear when v2 is out, watch this repo on GitHub: Watch → Custom → Releases.
Status
Draft 0.1. The spec may still change before 1.0.
Contributing
Open an issue or a pull request. If you change a schema, update SPEC.md andpatterns/starter.json in the same pull request and run the check above.
License
MIT, see LICENSE.
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi