AETumi MCP is expanding across the Model Context Protocol ecosystem, making AETumi's Three.js, WebGL and interactive 3D web resources easier to discover from compatible AI coding tools. Developers using Claude Code, Cursor and Codex can connect a single MCP server and pull production-ready 3D web components, scenes and full website templates directly into the build.
The announcement
AETumi MCP now runs as a hosted Model Context Protocol server over Streamable HTTP and is listed across the open MCP ecosystem. The goal is simple: when a developer's AI assistant needs real, production-ready 3D web resources, it can reach them without leaving the editor. That means interactive Three.js and WebGL scenes, React Three Fiber components, and full 3D website templates become part of the assistant's working context.
Why AETumi built an MCP server
Building interactive 3D on the web usually means switching between a browser, a component library and the code editor: copying snippets, losing context, and rebuilding the same renderer, scene graph and disposal logic every time. AETumi built its MCP server to close that gap. Instead of leaving the editor, an AI coding assistant can discover real AETumi resources and implementation guidance, while the underlying architecture stays visible and owned rather than hidden inside generated output.
What developers can discover
Through the MCP, an assistant can search and retrieve production-oriented 3D web resources:
- Interactive Three.js and WebGL hero sections and scenes
- 3D product viewers and configurators built with React Three Fiber
- Shader and gradient backgrounds for landing pages
- Scroll-driven 3D storytelling patterns
- Full 3D website templates across industries such as fashion, automotive, real estate and e-commerce
Each result points to a live, interactive reference, and the editable source is something you own after licensing rather than a black box.
What designers and teams get
The MCP is not only for engineers. Four experience-discovery tools speak the language of design and product, not Three.js internals: you can find experiences by industry (beauty, automotive, fashion, real estate, e-commerce), by visual style (dark cinematic, editorial luxury, glass luxury, bright architectural), or by business goal (product launch, premium e-commerce, property presentation). A single natural-language query can combine all three.
How it works with AI coding agents
Once connected, the assistant calls intent-based tools: searching the catalog, filtering experiences by industry, style or goal, and recommending a production build approach with Three.js / WebGL architecture, framework choice and performance guidance. The developer stays in Claude Code, Cursor or Codex; the 3D web context comes to the task instead of the other way around. Recommendations keep the essentials in view: a single renderer/scene owner, a capped device pixel ratio, one animation loop, off-screen pausing, reduced-motion fallbacks and strict resource disposal.
Example requests it answers
- “Find me a Three.js hero section for a luxury product website.”
- “Show me WebGL shader backgrounds for a fashion landing page.”
- “Find a 3D product viewer built with React Three Fiber.”
- “Recommend a build for a 3D headphone website.”
- “Show me cinematic scroll animations for a product launch.”
Availability
AETumi MCP is available through the AETumi hosted endpoint, on GitHub, and across selected MCP registries in the ecosystem. Connecting takes one line:
claude mcp add --transport http aetumi https://mcp.aetumi.appExplore AETumi MCP or read the source on GitHub.
What's next
AETumi continues to expand the library and the resources reachable through the MCP: more industry experiences, more Three.js and WebGL patterns, and deeper documentation for connecting from AI coding tools. The direction stays the same: keep production 3D web discoverable from the assistant, and keep developers in control of architecture, performance and accessibility.
Frequently asked questions
What is AETumi MCP?
A Model Context Protocol server that lets AI coding assistants discover and use production-ready Three.js, WebGL and React Three Fiber components, 3D website templates and interactive scenes from the AETumi library.
How do I connect it?
Run claude mcp add --transport http aetumi https://mcp.aetumi.app. It is hosted, so there is nothing to install or build locally.
Does it write my code?
No. It brings real 3D web resources and implementation guidance into your assistant; you keep full control of architecture, performance and the editable source you own.
Building an interactive 3D website? Connect AETumi MCP and bring real Three.js and WebGL resources into your AI coding workflow, from AETumi, the AI-native premium 3D web platform.