AI & Skills
Use RilayKit documentation with AI agents. Install the RilayKit skill for Claude Code, Cursor, and other AI coding assistants.
AI & Skills
RilayKit is designed to be AI-friendly. Use the documentation directly with AI agents, or install the RilayKit skill for context-aware assistance.
LLM-Ready Documentation
The full RilayKit documentation is available in machine-readable formats for AI agents:
| Endpoint | Description |
|---|---|
/llms.txt | Lightweight index of all pages with titles and descriptions |
/llms-full.txt | Complete documentation content in plain markdown |
Use these endpoints to feed RilayKit knowledge into any AI tool that supports llms.txt.
Install the RilayKit Skill
The RilayKit skill gives AI coding agents deep knowledge of the framework: builder patterns, validation, conditions, hooks, and real-world patterns.
npx skills add andyoucreate/rilaykit -a claude-codenpx skills add andyoucreate/rilaykit --allnpx skills add andyoucreate/rilaykit -g --allThe skill uses progressive disclosure — only the relevant reference is loaded into context when needed, keeping the AI's context window efficient.
What's Included
The skill covers the entire RilayKit API across three reference files:
Core
ril instance, component registry, Standard Schema validation, conditions with when() builder
Forms
Form builder API, row layouts, components, field hooks, form hooks, condition hooks
Workflow
Flow builder, navigation, step data helpers, persistence, analytics, real-world patterns
Supported Agents
The skill works with any agent compatible with the skills.sh ecosystem, including:
- Claude Code — Anthropic's CLI agent
- Cursor — AI-powered code editor
- Windsurf — Codeium's AI IDE
- OpenCode — Open-source coding agent
- And 40+ more agents
Verify Installation
After installing, ask your AI agent:
How do I create a multi-step workflow with conditional steps using RilayKit?The agent should reference the @rilaykit/workflow builder API with when() conditions — not generic form library patterns.