modelcontextprotocol/ext-apps

What it solves

MCP Apps allows developers to move beyond simple text and structured data responses from AI tools. It provides a standardized way to embed interactive user interfaces—such as charts, forms, dashboards, and design canvases—directly within the chat interface of compliant AI clients like Claude and ChatGPT.

How it works

It extends the Model Context Protocol (MCP) by allowing tools to declare UI resources using a ui:// URI. When an LLM calls a tool, the host client fetches the HTML interface and renders it in a sandboxed iframe. The system supports bidirectional communication: the host passes data to the UI via notifications, and the UI can trigger other tools through the host.

Who it’s for

• App Developers: Those building interactive Views using the provided SDKs (including React hooks).
• Host Developers: Those embedding these interactive Views into their own chat clients.
• MCP Server Authors: Those wanting to register tools with UI metadata to enhance the user experience.

Highlights

• Broad Client Support: Compatible with Claude, ChatGPT, VS Code, Goose, and Postman.
• Agent-Driven Development: Includes "Agent Skills" that allow AI coding agents to scaffold, migrate, or add UI to MCP servers automatically.
• Framework Agnostic: Provides starter templates for React, Vue, Svelte, Preact, Solid, and Vanilla JS.
• Rich Examples: Includes a wide range of demo apps, from 3D globe viewers and real-time GLSL shader renderers to customer retention heatmaps and PDF viewers.