Next Gen UI Agent – open-source engine for AI-generated UI

UI Agent uses LLMs to generate safe UI components that visualize rich backend data shaped for user prompt.

The problem UI Agent solves is: how to intelligently visualize rich backend data in AI assistant GUIs?

In upcoming releases, the agent will also maintain UI state and layouts to keep UI flow consistent across chats, and add other capabilities. Stay tuned ;-)

Why use Next Gen UI Agent?

Rich user experience – Extend text-based LLM chats with AI-generated GUI components such as cards, tables, charts, video players, and image galleries.
Secure UI rendering – LLM doesn't generate UI code, but declarative component descriptions, rendered using secure native renderers
Framework agnostic – Choose the AI framework/protocol and UI component framework you prefer.
AI framework integration – Seamless integration with MCP and A2A and others.

Example of a rich card component that includes an image and values from structured data:

Card UI Component

How to use Next Gen UI Agent

Start with the Architecture guide to see how to integrate the agent into your AI assistant.
Check the LangGraph & Web Components quickstart.
Run the LangGraph Movies assistant example to see the UI agent in action.