Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
docs(huggingface): index the Hugging Face article, avatar rigs, and Space
nirholaseeb0d805fffe
diff --git a/blog/index.html b/blog/index.htmlindex 1450d1b12..2ce869fd0 100644--- a/blog/index.html+++ b/blog/index.html@@ -537,6 +537,11 @@<span class="bl-press-title">Embodied, Intelligent, and Self-Custodied: three.ws and the 3D AI Agent Stack</span><span class="bl-press-arrow">↗</span></a>+ <a class="bl-press-card" href="https://huggingface.co/blog/three-ws/giving-ai-agents-bodies-and-wallets" target="_blank" rel="noopener">+ <span class="bl-press-outlet">Hugging Face</span>+ <span class="bl-press-title">Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws</span>+ <span class="bl-press-arrow">↗</span>+ </a></div></section></main>diff --git a/data/pages.json b/data/pages.jsonindex 53ade1722..a9104bb4b 100644--- a/data/pages.json+++ b/data/pages.json@@ -4540,6 +4540,13 @@"priority": 0.6,"changefreq": "monthly"},+ {+ "path": "/docs/huggingface",+ "title": "Docs \u00b7 three.ws on Hugging Face",+ "description": "Hugging Face is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under its organization account, and this is the index of the article, the avatar rigs, and the Space.",+ "priority": 0.6,+ "changefreq": "monthly"+ },{"path": "/docs/aws-builder-center","title": "Docs · three.ws on the AWS Builder Center",diff --git a/data/rss/items.json b/data/rss/items.jsonindex 7681e67d4..fc460d536 100644--- a/data/rss/items.json+++ b/data/rss/items.json@@ -32,6 +32,24 @@"rpc"]},+ {+ "id": "huggingface-bodies-and-wallets",+ "slug": "huggingface-bodies-and-wallets",+ "title": "Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws",+ "date": "2026-07-13T12:00:00.000Z",+ "author": "three.ws",+ "link": "https://huggingface.co/blog/three-ws/giving-ai-agents-bodies-and-wallets",+ "summary": "Our community article on Hugging Face, the platform where the machine-learning world publishes and checks models: four problems an embodied agent has to solve, and the shipped surface that answers each one. Published alongside a repository of rigged avatars and a running Space that renders them in the browser.",+ "body_html": "<p><strong>Most AI agents are a chat window. Ours walk around and pay each other.</strong> That is the opening line of our community article on Hugging Face, and the rest of it is the argument for why the four hard parts belong in one stack instead of four products.</p>\n<h2>What the article covers</h2>\n<ul>\n<li><strong>The body problem:</strong> text or a photo becomes a textured mesh, then a humanoid skeleton, then retargeted animation that plays on a rig nobody hand-authored.</li>\n<li><strong>The world problem:</strong> those bodies inhabit persistent multiplayer scenes rather than sitting alone in a single embed.</li>\n<li><strong>The wallet problem:</strong> agents hold real keys and settle per call in USDC over HTTP 402, so buying a capability is a request rather than a signup form.</li>\n<li><strong>The brain problem:</strong> an LLM that calls tools, keeps memory, and drives expression on the mesh in real time.</li>\n</ul>\n<p>It closes on a two-minute path from reading to a generated avatar of your own. Published next to it on the same account: <a href=\"https://huggingface.co/three-ws/avatars\" rel=\"noopener\">a repository of rigged avatars</a> as plain glTF under the MIT license, and <a href=\"https://huggingface.co/spaces/three-ws/avatar-viewer\" rel=\"noopener\">a running Space</a> that loads them in an interactive 3D viewer, so the claim that the rigs are animation-ready can be checked in the browser in the same sitting.</p>\n<p>Read it: <a href=\"https://huggingface.co/blog/three-ws/giving-ai-agents-bodies-and-wallets\" rel=\"noopener\">huggingface.co/blog/three-ws/giving-ai-agents-bodies-and-wallets</a></p>",+ "tags": [+ "huggingface",+ "3d",+ "agents",+ "x402",+ "agent-payments",+ "blog"+ ]+ },{"id": "aws-builder-center-3d-agents-onchain-payments","title": "three.ws Publishes \"Build Autonomous AI Agents with 3D Bodies and On-Chain Payments\" on the AWS Builder Center",diff --git a/docs/huggingface.md b/docs/huggingface.mdnew file mode 100644index 000000000..1684babd2--- /dev/null+++ b/docs/huggingface.md@@ -0,0 +1,59 @@+# three.ws on Hugging Face++[Hugging Face](https://huggingface.co) is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under the organization account [**@three-ws**](https://huggingface.co/three-ws): a model repository of rigged avatars, a running Space that renders them in the browser, and community articles about the stack behind them.++This page is the index of that presence: what is published, where it lives, and what code in this repository each piece documents. If you are adding an article or a model repo, the checklist at the bottom is the process.++- **Profile:** [huggingface.co/three-ws](https://huggingface.co/three-ws)+- **Account:** organization account, byline "three.ws"+- **Related surfaces:** [the avatar pipeline](./avatar-pipeline.md), [the `<agent-3d>` web component](./web-component.md), [agent wallets](./agent-wallets.md), [x402](./autonomous-x402.md)++## Published articles++### Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws++Published 13 July 2026 as a Hugging Face community article. It opens on the thesis the whole platform runs on: most AI agents are a chat window, and ours walk around and pay each other.++> [Read it on Hugging Face](https://huggingface.co/blog/three-ws/giving-ai-agents-bodies-and-wallets)++Four problems in order, each answered with a shipped surface: **the body problem** (text or a photo becomes a textured mesh, then a humanoid skeleton, then retargeted animation), **the world problem** (those bodies inhabit persistent multiplayer scenes instead of a single embed), **the wallet problem** (agents hold real keys and settle per call in USDC over HTTP 402), and **the brain problem** (an LLM that calls tools, holds memory, and drives expression on the mesh). It closes with a two-minute try-it path and the argument for why the four belong in one stack rather than four products.++- **Surfaces it covers:** [/forge](https://three.ws/forge), [/play](https://three.ws/play), [/agents-live](https://three.ws/agents-live), [the x402 manifest](https://three.ws/.well-known/x402.json)+- **Code it documents:** [the avatar pipeline](./avatar-pipeline.md), [animation retargeting](./animations.md), [agent wallets](./agent-wallets.md), [autonomous x402](./autonomous-x402.md)++## Published models++### three-ws/avatars++Rigged, animation-ready avatars from the platform, published as plain uncompressed glTF binaries under the MIT license so they load in any standard viewer without a Draco or Meshopt decoder.++> [Open the model repository](https://huggingface.co/three-ws/avatars)++The repository carries a sample rigged character, an avatar generated from a text prompt through the Forge TRELLIS lane, and a reference humanoid skeleton of the kind [`src/glb-canonicalize.js`](https://github.com/nirholas/three.ws/blob/main/src/glb-canonicalize.js) maps onto the canonical bone set before [`src/animation-retarget.js`](https://github.com/nirholas/three.ws/blob/main/src/animation-retarget.js) drives it. Hugging Face renders every file in its built-in 3D viewer, so the rigs are inspectable from the page.++Note that the files here are deliberately uncompressed. Everything three.ws serves in production is Meshopt or Draco compressed for load time; these copies trade bytes for the widest possible viewer compatibility.++## Published Spaces++### three.ws Avatar Viewer++> [Open the Space](https://huggingface.co/spaces/three-ws/avatar-viewer)++A running Space that loads the avatars above in an interactive 3D viewer, so a reader who arrives from the article can orbit a real rig without leaving Hugging Face or installing anything. The equivalent on our own domain is [the avatar inspector](./avatar-inspector.md).++## Publishing checklist++Hugging Face community articles accept Markdown, so a draft in `docs/` is the article. To take one from draft to published:++1. **Write the draft as `docs/huggingface-<topic>.md`.** Every code sample must be real code from this repository, and every claim must be verifiable against it. This audience reads the linked source and will open the model files.+2. **Check the rules:** `npm run check:rules -- --paths docs/huggingface-<topic>.md`.+3. **Check the links:** `npm run audit:docs`.+4. **Publish** from [huggingface.co/new-blog](https://huggingface.co/new-blog) using the organization account, pasting the Markdown body.+5. **Record the canonical URL** by adding a section to this page: title, publish date, the live URL, the surfaces it covers, and the code it documents.+6. **Announce it** by adding an item to `data/rss/items.json` (the curated news feed) with the canonical Hugging Face URL as `link`, then regenerate with `npm run build:news`. The item becomes a page under `/news/<slug>` and enters the RSS feed automatically.+7. **Add the press card** to the "In the press" grid at the bottom of `blog/index.html`, so the article is reachable from [/blog](https://three.ws/blog).+8. **Log it** in `data/changelog.json` with the `docs` tag.++## Why we publish here++Hugging Face is where people go to check whether a model claim is real, and the check is cheap: open the repo, load the file, look at the rig. A marketing page says an avatar is animation-ready; a glTF with a named humanoid skeleton that plays a retargeted clip in the browser's own viewer proves it. Publishing the artifacts next to the article means a reader can verify the body problem is actually solved in the same sitting they read about it, then follow the same thread into [the wallet](./agent-wallets.md) and [the payments](./autonomous-x402.md).diff --git a/docs/nav.json b/docs/nav.jsonindex 93a66114a..b6b392510 100644--- a/docs/nav.json+++ b/docs/nav.json@@ -967,6 +967,10 @@"label": "NVIDIA Inception","path": "nvidia-inception"},+ {+ "label": "Hugging Face",+ "path": "huggingface"+ },{"label": "AWS Builder Center","path": "aws-builder-center"diff --git a/docs/start-here.md b/docs/start-here.mdindex ca8ec8c3a..2507becc1 100644--- a/docs/start-here.md+++ b/docs/start-here.md@@ -170,6 +170,7 @@ Deeper references that don't fit a track above but answer real questions:- [Big-tech recognition dispatch board](./big-tech-recognition-week.md): every big-tech surface ranked against every other by the odds of a published mention this week, which six are already one action from shipping, and what to stop chasing- [The generator was never the hard part](./nvidia-nemotron-spotlight.md): our Nemotron Nano write-up, published on the NVIDIA Developer Forums- [three.ws on the AWS Builder Center](./aws-builder-center.md): the index of our published AWS engineering writing, what code each article documents, and the checklist for publishing the next one+- [three.ws on Hugging Face](./huggingface.md): our organization account there, the community article on giving agents bodies and wallets, the published avatar rigs and the Space that renders them, and the checklist for publishing the next one- [The onboarding tier](./onboarding-tier.md): why a first-time visitor sees ~20 nav destinations instead of ~100, how the Simple ⇄ Everything switch works, and which tier a new page belongs in- [What an agent can do](./agent-abilities/ABILITIES.md): the full abilities dossier, generated from the source article- [Use cases and example workflows](./content/use-cases/README.md): five audience-specific walkthroughs tying shipped features to concrete outcomesdiff --git a/public/features.json b/public/features.jsonindex c46382d5b..3b7949027 100644--- a/public/features.json+++ b/public/features.json@@ -1,5 +1,5 @@{- "generated_at": "2026-08-22T02:40:42.481Z",+ "generated_at": "2026-08-24T15:23:37.938Z","generated_by": "scripts/build-page-index.mjs from data/pages.json","site": {"name": "three.ws",@@ -5887,6 +5887,16 @@"tags": [],"showcase": false},+ {+ "path": "/docs/huggingface",+ "title": "Docs · three.ws on Hugging Face",+ "description": "Hugging Face is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under its organization account, and this is the index of the article, the avatar rigs, and the Space.",+ "added": null,+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },{"path": "/docs/aws-builder-center","title": "Docs · three.ws on the AWS Builder Center",@@ -7315,6 +7325,16 @@"tags": [],"showcase": false},+ {+ "path": "/news/huggingface-bodies-and-wallets",+ "title": "Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws",+ "description": "Our community article on Hugging Face, the platform where the machine-learning world publishes and checks models: four problems an embodied agent has to solve, and the shipped surface that answers each one. Published alongside a repository of rigged avatars and a running Space that renders them in the browser.",+ "added": null,+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },{"path": "/news/aws-builder-center-3d-agents-onchain-payments","title": "three.ws Publishes \"Build Autonomous AI Agents with 3D Bodies and On-Chain Payments\" on the AWS Builder Center",diff --git a/public/llms-full.txt b/public/llms-full.txtindex 7ba3e1241..afca1476d 100644--- a/public/llms-full.txt+++ b/public/llms-full.txt@@ -3563,6 +3563,12 @@ URL: https://three.ws/docs/ibm-x402-mcpA menu of IBM AI capabilities (chat, code, embeddings, document analysis, forecasting) that any agent or MCP client can use by paying a few cents per call, with no account and no subscription.+### Docs · three.ws on Hugging Face++URL: https://three.ws/docs/huggingface++Hugging Face is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under its organization account, and this is the index of the article, the avatar rigs, and the Space.+### Docs · three.ws on the AWS Builder CenterURL: https://three.ws/docs/aws-builder-center@@ -4430,6 +4436,12 @@ URL: https://three.ws/news/quicknode-startup-programthree.ws has been accepted into the Quicknode Startup Program and approved for free infrastructure credits: more RPC headroom for the Solana-first agent economy, provider redundancy, and Quicknode's founder network.+### Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws++URL: https://three.ws/news/huggingface-bodies-and-wallets++Our community article on Hugging Face, the platform where the machine-learning world publishes and checks models: four problems an embodied agent has to solve, and the shipped surface that answers each one. Published alongside a repository of rigged avatars and a running Space that renders them in the browser.+### three.ws Publishes "Build Autonomous AI Agents with 3D Bodies and On-Chain Payments" on the AWS Builder CenterURL: https://three.ws/news/aws-builder-center-3d-agents-onchain-paymentsdiff --git a/public/llms.txt b/public/llms.txtindex 1445523fa..531adfbce 100644--- a/public/llms.txt+++ b/public/llms.txt@@ -628,6 +628,7 @@ Docs, tutorials, and reference.- [Docs · The $THREE micro-buy loop](https://three.ws/docs/three-microbuy): Many tiny, real on-chain buys of $THREE per minute, each triggered by and paid through a settled pay-per-call payment. It exists to put continuous, verifiable buy pressure behind the platform's own token.- [Docs · IBM watsonx and Granite](https://three.ws/docs/ibm): three.ws agents can think on IBM Granite foundation models served through IBM watsonx.ai. The whole integration for developers and operators: model selection, credentials, request shapes, and the fallback behind it.- [Docs · IBM Granite pay-per-call MCP](https://three.ws/docs/ibm-x402-mcp): A menu of IBM AI capabilities (chat, code, embeddings, document analysis, forecasting) that any agent or MCP client can use by paying a few cents per call, with no account and no subscription.+- [Docs · three.ws on Hugging Face](https://three.ws/docs/huggingface): Hugging Face is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under its organization account, and this is the index of the article, the avatar rigs, and the Space.- [Docs · three.ws on the AWS Builder Center](https://three.ws/docs/aws-builder-center): The AWS Builder Center is Amazon's community publishing platform for engineers building on AWS. three.ws publishes there under its official organisation account, and this is the index of what has been published.- [Docs · three.ws on AWS Marketplace](https://three.ws/docs/aws-marketplace): three.ws is a verified AWS Partner. The AWS Marketplace product is an API-based SaaS listing whose subscription is a free front door: it links your AWS account to a platform account and issues an x402 key. Integration, endpoints, and current listing status.- [Docs · three.ws on AWS](https://three.ws/docs/aws-partner-spotlight): Most enterprise AI agents live inside a chat box. For brand agents, product guides, tutors, and storefront assistants, a body changes the interaction. How three.ws builds that on AWS.@@ -767,6 +768,7 @@ Product launches, integrations, and announcements from three.ws.- [News & Announcements](https://three.ws/news/): Product launches, integrations, and announcements from three.ws.- [How to Monetize an MCP Server: Paid Tools with x402](https://three.ws/news/monetize-mcp-server-x402-paid-tools): The seller-side guide to agent payments: choose work worth charging for, price it in atomic USDC units, answer unpaid calls with a real 402, verify then work then settle, charge per tool inside an MCP tools/call, and get indexed where agents shop for paid services.- [three.ws Joins the Quicknode Startup Program](https://three.ws/news/quicknode-startup-program): three.ws has been accepted into the Quicknode Startup Program and approved for free infrastructure credits: more RPC headroom for the Solana-first agent economy, provider redundancy, and Quicknode's founder network.+- [Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws](https://three.ws/news/huggingface-bodies-and-wallets): Our community article on Hugging Face, the platform where the machine-learning world publishes and checks models: four problems an embodied agent has to solve, and the shipped surface that answers each one. Published alongside a repository of rigged avatars and a running Space that renders them in the browser.- [three.ws Publishes "Build Autonomous AI Agents with 3D Bodies and On-Chain Payments" on the AWS Builder Center](https://three.ws/news/aws-builder-center-3d-agents-onchain-payments): Our second article on the official AWS Builder Center: how to build browser-native 3D AI agents with memory, on-chain wallets, and autonomous x402 payments, available through AWS Marketplace and embeddable in one line of HTML.- [AWS Publishes "The Agentic Economy Is Here", a Builder's Guide Built on the three.ws Stack](https://three.ws/news/aws-agentic-economy-three-ws): AWS Strategic Partnerships published a full builder's guide to the agentic economy on the AWS Builder Center, and it uses the three.ws stack as its worked example throughout: the x402 payment handshake, 38 MCP servers, agent-to-agent payments, on-chain agent identity, agent tokens, and 3D embodiment. Written by AWS, not by us.- [Text to 3D Is Live - Type a Prompt, Download a 3D Model](https://three.ws/news/text-to-3d-is-live): Forge is live at three.ws/forge: type a description, get a textured, interactive 3D model you can orbit, view in AR, and download as a GLB. Image to 3D from photos works too - and agents can call the same pipeline over MCP.diff --git a/public/sitemap/index.html b/public/sitemap/index.htmlindex d16646515..9992a41b2 100644--- a/public/sitemap/index.html+++ b/public/sitemap/index.html@@ -83,7 +83,7 @@<main class="sm-wrap"><div class="sm-hero"><h1>Sitemap</h1>- <p>All 831 pages on three.ws, grouped by purpose. Looking for the machine-readable versions?</p>+ <p>All 833 pages on three.ws, grouped by purpose. Looking for the machine-readable versions?</p><div class="sm-formats"><a href="/sitemap.xml"><code>sitemap.xml</code> · for search engines</a><a href="/llms.txt"><code>llms.txt</code> · for AI agents</a>@@ -95,7 +95,7 @@</div><div class="sm-filter"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" aria-hidden="true"><circle cx="11" cy="11" r="7"/><path d="m21 21-4.3-4.3"/></svg>- <input id="sm-filter-input" type="search" placeholder="Filter 831 pages by name, path, or description…" aria-label="Filter pages" autocomplete="off" spellcheck="false" />+ <input id="sm-filter-input" type="search" placeholder="Filter 833 pages by name, path, or description…" aria-label="Filter pages" autocomplete="off" spellcheck="false" /><kbd aria-hidden="true">/</kbd></div><p class="sm-filter-count" id="sm-filter-count" role="status" aria-live="polite"></p>@@ -2276,7 +2276,7 @@<section class="sm-section" id="learn"><header>- <h2>Learn<span class="sm-count" data-count>371</span></h2>+ <h2>Learn<span class="sm-count" data-count>372</span></h2><p class="sm-section-desc">Docs, tutorials, and reference.</p></header><ul class="sm-list">@@ -4331,6 +4331,13 @@<span class="sm-desc">A menu of IBM AI capabilities (chat, code, embeddings, document analysis, forecasting) that any agent or MCP client can use by paying a few cents per call, with no account and no subscription.</span></a></li>+ <li>+ <a href="/docs/huggingface">+ <span class="sm-title">Docs · three.ws on Hugging Face</span>+ <span class="sm-path">/docs/huggingface</span>+ <span class="sm-desc">Hugging Face is where the machine-learning community publishes models, demos, and engineering writing. three.ws publishes there under its organization account, and this is the index of the article, the avatar rigs, and the Space.</span>+ </a>+ </li><li><a href="/docs/aws-builder-center"><span class="sm-title">Docs · three.ws on the AWS Builder Center</span>@@ -5322,7 +5329,7 @@<section class="sm-section" id="news"><header>- <h2>News<span class="sm-count" data-count>111</span></h2>+ <h2>News<span class="sm-count" data-count>112</span></h2><p class="sm-section-desc">Product launches, integrations, and announcements from three.ws.</p></header><ul class="sm-list">@@ -5347,6 +5354,13 @@<span class="sm-desc">three.ws has been accepted into the Quicknode Startup Program and approved for free infrastructure credits: more RPC headroom for the Solana-first agent economy, provider redundancy, and Quicknode's founder network.</span></a></li>+ <li>+ <a href="/news/huggingface-bodies-and-wallets">+ <span class="sm-title">Giving AI Agents 3D Bodies, Real Jobs, and Wallets on three.ws</span>+ <span class="sm-path">/news/huggingface-bodies-and-wallets</span>+ <span class="sm-desc">Our community article on Hugging Face, the platform where the machine-learning world publishes and checks models: four problems an embodied agent has to solve, and the shipped surface that answers each one. Published alongside a repository of rigged avatars and a running Space that renders them in the browser.</span>+ </a>+ </li><li><a href="/news/aws-builder-center-3d-agents-onchain-payments"><span class="sm-title">three.ws Publishes "Build Autonomous AI Agents with 3D Bodies and On-Chain Payments" on the AWS Builder Center</span>