three.ws: Browser-Native 3D AI Agents via Three.js and MCP

Repositorynirholas/three.wsTool

three.ws is an open-source platform designed to create and embed interactive 3D agents directly in the browser. It allows developers to turn text, images, or sketches into rigged GLB avatars, attach LLM reasoning via the Model Context Protocol (MCP), and embed them anywhere using a single Web Component.

Key points

  1. Generates 3D GLB models from text, images, or sketches through the 'Forge' tool, or accepts uploaded models.
  2. Equips avatars with LLM reasoning, memory, and remote MCP server capabilities.
  3. Enables zero-plugin browser embedding using native Web Components and Three.js.
  4. Integrates Web3 functionality with Solana wallet support, x402 micropayments, and the $THREE token.
  5. Released under the Apache-2.0 licence with public npm packages (@three-ws/sdk and @three-ws/mcp-server).

Why it matters

Most AI agents remain confined to simple text and audio interfaces. three.ws lowers the barrier for embodied AI on the web by combining procedural 3D generation, standardized Web Components, and MCP, allowing developers to deploy animated avatars without complex game engine setups.

Analysis

three.ws tackles the friction of building interactive 3D characters, which previously required heavy game engines, native clients, or dedicated 3D artists. By unifying 3D generation, Three.js rendering, and LLM orchestration into a single browser-native workflow, it makes spatial avatars accessible via simple HTML tags. The stack relies on WebGL, WebXR, GLTF/GLB, Web Components, and MCP. The target audience includes front-end developers, AI agent builders, and spatial web experimenters. In terms of maturity, three.ws is an early-stage but functional open-source project (224 stars, 61 forks) with published npm packages and clear contribution guidelines. Its close ties to Solana and memecoin platforms ($THREE on pump.fun) suggest a Web3-leaning, community-driven experiment that is still evolving into an enterprise-ready framework.

Found in

Summary generated by AI (Gemini 3.8 Flash, 6 October 2026). It may contain errors – the original source is authoritative.

Similar in content, found by the AI search.

  • Link:three.ws

    three.ws: 3D AI Agent Platform with Text-to-3D and Web Embedding

    three.ws is a web platform for creating and embedding animated 3D AI agents. Users can generate 3D models or avatars from text prompts or selfies, customise their behaviour, and embed them via web components directly into websites.

    AI & LLMs· Tool

  • Repository:img2threejs/img2threejs

    img2threejs: Procedural Code-Only Image-to-3D for Three.js

    img2threejs turns reference images into procedural, animation-ready Three.js models generated purely as code, avoiding traditional mesh extraction or photogrammetry.

    15.1kstarsPython

    AI & LLMs· Tool

  • Repository:mshumer/Claude-of-Duty

    Claude of Duty: Fully Procedural Browser FPS Built by AI Agents Using Three.js

    Claude of Duty is an open-source first-person shooter running in the browser on Three.js r180 and WebGL2. Spanning roughly 55,000 lines of code across 11 subsystems, it was constructed by an orchestrated fleet of AI agents. The game contains zero pre-baked art assets: every texture, 3D mesh, animation, and sound effect is generated procedurally from code at runtime.

    3.4kstarsJavaScript

    AI & LLMs· Demo

  • Repository:Podcast72/showroom-la-finestra-webmcp

    WebMCP: In-Browser Agent Interface Layer via document.modelContext

    The showroom-la-finestra-webmcp repository demonstrates how existing websites can expose structured business knowledge and safe actions directly to AI agents without running a separate MCP server. By using an in-browser layer via `document.modelContext`, the website explains itself rather than relying on DOM scraping.

    1starsJavaScript

    AI & LLMs· Resource

  • Repository:ashemag/human-atlas

    Human Atlas: Open-Source 3D Anatomy Explorer in the Browser

    Human Atlas is an open-source, interactive 3D anatomy explorer built with React, Three.js, and shadcn/ui. It breaks down the BodyParts3D reference model into 2,234 selectable meshes across 15 anatomical systems, offering search, layer toggles, and exploded views.

    2.9kstarsTypeScript

    Web development· Tool

Let's talk about your project

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.