Outkit transforms raw AI text responses into structured, interactive UI components. Instead of showing users walls of markdown, Outkit automatically detects tables, charts, checklists, alerts, and 50+ other component types in your AI output — and returns structured data your frontend can render beautifully. One line of code. 10x better AI output.Documentation Index
Fetch the complete documentation index at: https://docs.outkit.dev/llms.txt
Use this file to discover all available pages before exploring further.
How It Works
Outkit is a post-processing layer. It doesn’t proxy your LLM calls or touch your tokens. Your agents, orchestration, and API keys stay untouched. At the very end, when you have a response meant for a human, you pass it through Outkit.- Your LLM generates a response — any model, any provider, any framework
- Outkit understands the response — detects structure like comparisons, metrics, steps, timelines
- Outkit extracts structured props — returns typed component data (table columns/rows, chart datasets, checklist items)
- Your frontend renders rich UI — using Outkit’s renderer or your own components
Get Started in One Command
The Outkit CLI is the recommended way to set up Outkit. It detects your framework, picks a profile, installs the SDK, writes the backend proxy, and prints the swap snippet — in one command.Ways to Integrate
CLI
Recommended. Wire your project, manage keys, sync design tokens, and tail logs from the
terminal.
React SDK
Drop-in React component with streaming, Shadow DOM isolation, and design tokens.
REST API
Direct HTTP integration for custom applications, backends, and AI pipelines.
Quick Links
Quickstart
Get running in under 5 minutes.
Components
Explore the 50+ component types Outkit can generate.
Examples
See real input/output examples.