Show HN: Stop returning raw JSON from MCP servers, build rich inline UIs(medium.com)
medium.com
Show HN: Stop returning raw JSON from MCP servers, build rich inline UIs
https://medium.com/towards-artificial-intelligence/mcp-apps-build-interactive-apps-directly-inside-your-ai-agents-chat-c571678099e3
7 コメント
What a terrible idea this ui protocol is. MCP is already pretty much “prompt injection as a service “. This creates a little-known side channel to make it easier to slip an exploit under people’s radar.
I get where you’re coming from, but there are some security practices in place. The host client renders views inside a strictly sandboxed `<iframe>`. Any action the UI wants to take must pass an auditable message back to the host application, which triggers an explicit user-permission prompt.
I stumbled onto this capability while using the Spotify MCP App inside Claude. a fully functional, interactive playlist an dmusic player widget spin up inline, instead of a standard markdown list of text links. It turns out this is built on the official MCP Apps spec extension that Anthropic and OpenAI standardized a few months back. It lets your server declare a ui:// URI, which the host grabs and mounts inside a sandboxed iframee. Core mcp blew up overnight, but this extension still feels pretty underground. It makes me wonder if we're heading toward a weird shift where AI agents become the primary browsers of the web, and front-end dev becomes about building micro-widgets for LLMs instead of humans.
Not to continue piling on here but if we want to shift from a human consumer to an LLM consumer, I think the WORST thing you can do is obscure raw data behind some kind of "widget"