Building Interactive UIs in VS Code with MCP Apps
summary
A live demo profiles a Go app running bubble sort and Fibonacci for five seconds using Go pprof and an MCP tool, then returns JSON plus a resource reference to a bundled HTML UI so VS Code fetches and renders an interactive flame graph as a sandboxed iframe inside the chat (not a link). The iframe calls back to the MCP server for fresh data, the UI updates in place, the pattern is used by Shopify and Excalidraw, and the build is scaffolded from the MCP repository and run via GitHub Copilot CLI.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:58What is MCP
- 2:24MCP components
- 3:26Early MCP limitations
- 4:29MCP apps for UIs
- 5:34MCP apps workflow
- 6:37MCP apps use cases
- 8:23Companies using MCP
- 9:46Live demo setup
- 11:10Flame graph MCP app
- 13:14Running the MCP app
- 14:38Security and sandboxing
- 15:20Closing remarks
related talks
26mBuild your First MCP App
38mEvolving AI chat with MCP Apps
29mMCP Apps: Primitives, discovery, and the Future of Software
28mBuilding MCP servers with VS Code — Level up your MCP
15mWhen chatbots grow buttons: Building MCP apps
01h 20mReal World Development with GitHub Copilot and VS Code
29mBuilding MCP Apps With GraphQL Patterns You Already Know
18mHow I shipped more code and products than ever before