core

Define schemas, catalogs, and prompts for generating JSON UI specs.

15.9k|859|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/vercel-labs/json-render --skill core-vercel-labs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: core
Source: https://github.com/vercel-labs/json-render/tree/main/skills/core
Command: npx skills add https://github.com/vercel-labs/json-render --skill core-vercel-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Core package provides a structured foundation for AI-driven UIs by enabling developers to define schemas, catalogs, and prompts that govern how models generate and stream UI specifications.

Core Features & Use Cases

  • Define schemas to describe the spec and catalog shapes.
  • Provide tooling to create catalogs that map components and actions for guardrails and consistency.
  • Offer spec streaming and generation utilities to produce JSON specs for UI and video generation, with validation and state integration.

Quick Start

Define a schema with defineSchema, create a catalog with defineCatalog, and generate prompts and specs using the core utilities.

Frequently Asked Questions about core

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I define schemas and catalogs for AI-driven UI generation?

You define schemas to describe spec shapes and build catalogs to map components and actions, establishing guardrails that ensure AI models produce predictable and valid JSON UI specifications.

What is the best way to enforce validation and guardrails for AI-generated UI specs?

The best way to enforce validation for AI-generated UI specs is to use a catalog that maps allowed components and actions, combined with schema validation to ensure safe, predictable outputs.

How do I stream JSON UI specifications from an AI model?

You stream JSON UI specifications by using a spec stream compiler utility, which generates and validates the JSON spec incrementally as the AI model produces the output.

Can I integrate state management with AI-generated UI specifications?

Yes, you can integrate state management with AI-generated UI specifications using a StateStore utility and prop resolution functions to bind dynamic state values into the generated UI components.

Do I need to define a catalog before generating prompts for AI UI generation?

Yes, you need to define a catalog before generating prompts, because the catalog provides the component mapping and constraints required to create accurate prompts and enforce guardrails during generation.