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
feat(awesome): publish Awesome 3D Agents, a curated list of the tools for giving an AI a body
nirholasa4aacda277b1
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex 5cb422375..5d9d8576f 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -6,12 +6,14 @@ Public history for [three.ws](https://three.ws), newest first. New pages come fr## 2026-09-07+- **Awesome 3D Agents** (`/awesome`): A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.- **Docs · SEO and crawler rendering** (`/docs/seo`): How a JavaScript site is served to search engines: the four server-side render paths, the contract that keeps a page and its sitemap entry agreeing, and how to audit the live origin.- **My Creations** (`/mine`): Everything you have made on three.ws in one newest-first grid: the agents you built, the 3D avatars you saved, and every model you forged, including models forged in this browser before you had an account. Open, edit, or copy a link to any of them in one click.- **3D generation and the whole site keep working when our asset storage does not**: Early on 2026-09-07 our object storage stopped accepting the platform's credential, and because every generation parks its reference image there, 3D generation stopped on the website and in ChatGPT at the same time, agent registration could not store its manifest, and images loaded through our CDN stopped resolving. Three things changed so that a storage fault is never again a blank page. Avatars, thumbnails and models now fall back to the bucket's public domain automatically, so the site keeps rendering. When generation genuinely cannot run you get a plain message saying storage is recovering and to try again shortly, instead of a raw error from the storage vendor. And our own health checks now watch the storage credential directly and treat a silent forge as an outage rather than a quiet hour, so we hear about this class of failure before you do. `[fix, infra]`- **A payment we cannot read now comes back with the price list, not a bare error**: When an agent pays one of our x402 endpoints and the payment header arrives mangled (truncated in transit, double-encoded, wrapped by a proxy), the answer used to be a plain 400 error with no prices in it, so the agent had nothing left to work from and no way to retry. Every rejected payment now returns the ordinary 402 with the full list of accepted rails, tokens and amounts attached, exactly as it does when a payment is simply missing. That is what the x402 spec and the OKX seller SDK both do, and it means a paying agent can always read the quote again and try once more. Only a real server fault answers without a quote. (`/docs/okx-marketplace`) `[fix, improvement]`- **A sharper cutout for images with hair, fur, or fine edges**: Background removal now offers BiRefNet as an optional model alongside the fast default. The default model smears wispy boundaries into a soft halo, and because every image-to-3D lane rebuilds geometry from the cut-out subject, that halo turned into real polygons on the finished mesh. BiRefNet keeps hair strands, fur, foliage, and thin straps intact. It takes about six seconds against the default's one, so it stays opt-in: pass model "birefnet" on the background-removal endpoint or pick it from the MCP tool when the edge is what matters. `[improvement, feature]`- **Archived news stories with emoji in their summary keep their search listing**: Google Search Console flagged an archived story page for unparsable structured data. The cause: the archive trims a publisher summary to a fixed length before serving it, and the cut landed in the middle of an emoji, leaving half a character in the page data that search engines read. Trimming is now character-aware, so it never splits an emoji, a flag, or an accented letter, and the page data is checked for half characters before it is written. Around one archived story in seven thousand carried the flaw; all of them serve a valid listing now. The same fix covers the story summaries shown in link previews and the shared page descriptions across the site. (`/markets/news`) `[fix, infra]`+- **Awesome 3D Agents: a curated list of everything you need to give an AI a body**: Giving an agent a body means assembling a pipeline out of a dozen separate fields, and the map of that pipeline did not exist anywhere. There is now one: 134 hand-picked projects across 13 sections, covering text-to-3D generation, photo and video capture, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, physics and networking, agent frameworks, datasets, free assets, and where to learn all of it. Each entry says in one sentence what the thing does and why you would reach for it instead of the alternative. The list is searchable and filterable by tag at /awesome, copyable as Markdown, and published in the standard awesome-list format on GitHub so it can be read, forked, and contributed to there. Every link is fetched and verified before it ships. (`/awesome`) `[feature, docs]`- **Blender tools now open three.ws avatars, which Blender alone cannot**: Most three.ws avatars are delivered in a compressed form that Blender flatly cannot read: point stock Blender at one and it refuses the file. That made the new Blender tools useless on exactly the models this platform produces. Compressed models are now unpacked automatically before Blender opens them, for both of the common compression schemes, with nothing extra to install and your original file left untouched. The answer tells you when it happened. `[fix, sdk]`- **Keyboard focus is visible again on every /play HUD button, and the AGI filter counts clear AA contrast**: Six buttons on the /play world HUD (friends, jobs, avatar, wardrobe, shop and buy) had lost their keyboard focus ring: the page never defined the shared focus-ring colour those buttons outline with, so the browser dropped the outline entirely and a keyboard player could not see where they were. The ring is back. On /agi, the count inside a selected filter chip measured 4.37:1 against its background at 11px, just under the 4.5:1 AA floor; it now uses the primary ink and clears it comfortably. (`/play`) `[fix, improvement]`- **My Creations: one button that takes you back to everything you made**: A creator told us the making part was intuitive and the finding part was not: they built an agent, closed the tab, and had no obvious way back to it. There is now a My creations button in the header, and it opens a single page holding everything you have made, newest first: your agents, the 3D avatars saved to your account, and every model you forged, with Open, Edit and Copy link on each one. Signed out it still finds the models forged in this browser, so work made before you had an account is not lost. Filter by type, search by name or prompt, and hover any card to spin it in 3D. The create wizard and the Forge now point here too, so you are told where your work lives at the moment you make it. (`/mine`) `[feature, improvement]`diff --git a/README.md b/README.mdindex ae4b7413a..500ea547b 100644--- a/README.md+++ b/README.md@@ -8752,6 +8752,7 @@ A map of every user-facing route. [`STRUCTURE.md`](STRUCTURE.md) maps each produ| **Integrations** | `/cz`, `/lobehub/iframe` | CZ demo, LobeHub plugin || **IBM / Granite** | `/galaxy`, `/ibm/x402-demo`, `/ibm/hello` | Granite on watsonx.ai: semantic agent galaxy, pay-per-call Granite demo, hello-world embed || **Docs** | `/docs`, `/docs/widgets` | Developer documentation |+| **Awesome list** | `/awesome` | Awesome 3D Agents: a curated, link-checked list of the tools for giving an AI a body || **Legal** | `/legal/privacy`, `/legal/tos` | Privacy policy and terms |---diff --git a/STRUCTURE.md b/STRUCTURE.mdindex d0e0db4e3..fdd66b66e 100644--- a/STRUCTURE.md+++ b/STRUCTURE.md@@ -207,6 +207,7 @@ bottom.| Escrowed knocks (pay-per-reply) | [contracts/knock-escrow/](contracts/knock-escrow) | In-repo | Anchor program making a knock's payment conditional instead of final: `knock` parks the price in a vault owned by the knock's own PDA, and only three things can happen to it. `answer` pays the owner minus a fee capped at 10% in the program itself, `refuse` refunds the sender in full and takes no fee, and `reclaim` refunds the sender once the reply window closes and can be called by anyone. Each `KnockRecord` snapshots the fee that was in force when it was made, so an authority cannot reprice money already in escrow, and it stores only the SHA-256 of the message and the reply, never their text. The trust-me version of this settles straight to the owner at [api/knock/](api/knock) (row above); this is the version that works between strangers. Both sides are reachable from a browser: senders escrow and crank their own refund from the door page, owners open their on-chain door, answer for payment or refuse for a refund from [/knock](pages/knock.html), all signed by their own wallet ([src/knock/escrow-checkout.js](src/knock/escrow-checkout.js)). Docs: [contracts/knock-escrow/README.md](contracts/knock-escrow/README.md), [docs/knock.md](docs/knock.md). Tests: [tests/knock-escrow-program.test.js](tests/knock-escrow-program.test.js), [tests/knock-escrow-pda.test.js](tests/knock-escrow-pda.test.js), [tests/api-knock-escrowed.test.js](tests/api-knock-escrowed.test.js), [tests/api-knock-escrow-sync.test.js](tests/api-knock-escrow-sync.test.js) || Cross-chain SDKs | [sdk/](sdk), [solana-agent-sdk/](solana-agent-sdk), [agent-payments-sdk/](agent-payments-sdk), [agent-protocol-sdk/](agent-protocol-sdk) | Published | Cross-chain agent SDKs || MCP integration | [mcp-server/](mcp-server), [mcp-bridge/](mcp-bridge), [packages/*-mcp/](packages) | 43 npm servers + 7 hosted remote (50 total, incl. the free `/api/mcp-studio` 3D Studio), all in the [MCP registry](https://registry.modelcontextprotocol.io/?q=io.github.nirholas). Canonical metadata: [prompts/store-submissions/_generated/mcp-listing-source.json](prompts/store-submissions/_generated/mcp-listing-source.json), generated from the manifests by [scripts/build-mcp-listing-source.mjs](scripts/build-mcp-listing-source.mjs) (`npm run audit:mcp-listing` fails when it drifts) | Model Context Protocol surface, full server list in [docs/mcp.md](docs/mcp.md). Searchable index of every tool: [pages/mcp-tools.html](pages/mcp-tools.html) `→ /mcp-tools`, rendered from the generated [public/mcp-catalog.json](public/mcp-catalog.json) ([scripts/build-mcp-catalog.mjs](scripts/build-mcp-catalog.mjs), `npm run build:mcp-catalog`, staleness gated by `npm run audit:mcp-catalog`). Tool safety annotations are verified against each handler by [scripts/audit-mcp-safety.mjs](scripts/audit-mcp-safety.mjs) (`npm run audit:mcp-safety`, rules in [scripts/lib/mcp-safety-check.mjs](scripts/lib/mcp-safety-check.mjs), guarded-file list shared with the golden snapshot via [scripts/lib/mcp-tool-sources.mjs](scripts/lib/mcp-tool-sources.mjs)); docs [docs/mcp-safety.md](docs/mcp-safety.md) `→ /docs/mcp-safety`. Trust policy: [api/mcp-policy.js](api/mcp-policy.js) `-> GET /api/mcp-policy` turns those verified labels into a ready-to-paste client allowlist (profiles strict/balanced/open x formats claude/vscode/json), surfaced by the builder panel on `/mcp-tools`. Connector-review readiness (every paid stdio tool answers a clean x402 challenge, and the reviewer entitlement lifts it): [scripts/audit-mcp-reviewer-path.mjs](scripts/audit-mcp-reviewer-path.mjs), `npm run audit:mcp-reviewer`. Tests: [tests/mcp-safety-audit.test.js](tests/mcp-safety-audit.test.js), [tests/mcp-catalog.test.js](tests/mcp-catalog.test.js), [tests/api/mcp-policy.test.js](tests/api/mcp-policy.test.js) |+| Awesome 3D Agents (curated list) | Source [data/awesome.json](data/awesome.json), generator [scripts/build-awesome.mjs](scripts/build-awesome.mjs) (`npm run build:awesome`, drift-gated by `npm run check:awesome`), outputs [awesome/README.md](awesome/README.md) + [public/awesome.json](public/awesome.json), page [pages/awesome.html](pages/awesome.html) `-> /awesome`, link checker [scripts/check-awesome-links.mjs](scripts/check-awesome-links.mjs) (`npm run awesome:links`) | Live | An awesome-list of everything needed to give an AI agent a body: text-to-3D, capture and splatting, avatars, rigging, motion, lipsync and voice, web renderers, asset pipelines, physics and networking, agent frameworks, datasets, free assets, learning. One JSON source produces both the GitHub-facing list and the searchable page, so they cannot disagree. Every url is fetched and classified (ok / moved / bot-filtered / broken) before it ships; the checker stays out of the deploy path so a third-party outage cannot redden someone else's build. Contributing: [awesome/CONTRIBUTING.md](awesome/CONTRIBUTING.md) || Agent Skills pack | [.agents/skills/](.agents/skills) → manifest [.agents/skills/SKILLS.md](.agents/skills/SKILLS.md), [.agents/skills/skills-pack.json](.agents/skills/skills-pack.json) | In-repo; bundled in the `three-ws-core` [plugin](.claude-plugin/marketplace.json) | 40 portable `SKILL.md` folders (Agent Skills open standard) for any Claude surface — 3d/creative (cross-platform-safe), wallet/payments, intel/trading. Regenerate manifest: `node scripts/build-skills-pack.mjs`. Docs: [docs/agent-skills.md](docs/agent-skills.md) || Claude Code plugin marketplace | [marketplace/](marketplace) → plugins [three-ws-3d](marketplace/plugins/three-ws-3d), [three-ws-developer](marketplace/plugins/three-ws-developer) | In-repo · installable via `/plugin marketplace add nirholas/three.ws` | 3D generation + rigged avatars inside Claude Code (free lane + paid x402 tools) and developer tooling for building on three.ws. Index: [marketplace/README.md](marketplace/README.md) || SNS naming + pay-by-name | [api/sns.js](api/sns.js), [api/sns-subdomain.js](api/sns-subdomain.js), [api/threews/subdomain.js](api/threews/subdomain.js), [api/x402/pay-by-name.js](api/x402/pay-by-name.js), [src/solana/sns-subdomain.js](src/solana/sns-subdomain.js), [pages/threews-claim.html](pages/threews-claim.html) | In-repo | `*.threews.sol` subdomain mint, x402 payments addressed by name. Env: `THREEWS_SOL_PARENT_SECRET_BASE58` |diff --git a/awesome/CONTRIBUTING.md b/awesome/CONTRIBUTING.mdnew file mode 100644index 000000000..a578afa36--- /dev/null+++ b/awesome/CONTRIBUTING.md@@ -0,0 +1,64 @@+# Contributing to Awesome 3D Agents++The list is generated. `awesome/README.md` is an output file, so an edit made+there is overwritten by the next build. Entries live in+[`data/awesome.json`](../data/awesome.json).++## Add an entry++1. Open `data/awesome.json` and find the section it belongs in.+2. Add an object to that section's `items` array:++```json+{+ "name": "glTF-Transform",+ "url": "https://github.com/donmccurdy/glTF-Transform",+ "description": "Read, edit, optimise, and validate glTF from Node or the CLI.",+ "tags": ["oss", "js", "cli"]+}+```++3. Regenerate and check:++```bash+npm run build:awesome # rewrites awesome/README.md and public/awesome.json+npm run awesome:links # fetches every url and reports anything unreachable+```++4. Commit `data/awesome.json` together with both generated files.++## What gets in++The bar is "a working engineer would be glad someone showed them this".++- **It has to be usable now.** A repo with no release, no docs, and no commits+ in two years is a bookmark, not a recommendation.+- **It has to earn its section.** Fifteen text-to-3D models that do the same+ thing help nobody. If a new entry beats an existing one on the same job, say+ so in the description, or replace the old one.+- **One sentence, under 260 characters.** Say what it does and why you would+ reach for it over the alternative. The build fails on longer descriptions.+- **No em-dash (U+2014) or en-dash (U+2013).** House rule across this repository.+ Use a period, a comma, a colon, or parentheses. A plain hyphen is fine. The+ build fails on either dash.+- **No marketing copy.** "Blazing fast next-generation platform" tells a reader+ nothing. "Single image to 3D in under a second on one GPU" does.+- **Working links only.** `npm run awesome:links` classifies each url as ok,+ moved, blocked (a bot filter on a live site, verify it by hand), or broken.+ A broken url fails the run.++## Tags++Tags drive the filter chips on [three.ws/awesome](https://three.ws/awesome).+Reuse an existing tag when one fits rather than inventing a synonym; a tag used+only once stays searchable but is not offered as a chip. Common ones:++`oss`, `service`, `free`, `spec`, `dataset`, `research`, `foundational`,+`python`, `js`, `cpp`, `rust`, `threejs`, `gltf`, `fast`, `edge`.++## Sections++A new section is a bigger change than a new entry. Propose it in an issue+first, with the three or more entries that would fill it. Sections are ordered+along the pipeline: generate, capture, embody, rig, animate, speak, render,+ship, connect.diff --git a/awesome/README.md b/awesome/README.mdnew file mode 100644index 000000000..fc35d71d1--- /dev/null+++ b/awesome/README.md@@ -0,0 +1,232 @@+# Awesome 3D Agents++[](https://awesome.re)++> Everything you need to give an AI agent a body.++A curated list of the models, engines, formats, and research that turn a language model into something you can see, animate, talk to, and put on a web page. Generation, rigging, motion, faces, voice, renderers, and the agent frameworks that drive them.++134 entries across 13 sections. Browsable, searchable, and filterable at [three.ws/awesome](https://three.ws/awesome). Every link is checked with `npm run awesome:links`.++## Contents++- [Text and image to 3D](#text-and-image-to-3d) (16)+- [Photos and video to 3D](#photos-and-video-to-3d) (14)+- [Avatars and characters](#avatars-and-characters) (8)+- [Rigging and retargeting](#rigging-and-retargeting) (6)+- [Motion and animation](#motion-and-animation) (9)+- [Faces, lipsync, and voice](#faces-lipsync-and-voice) (15)+- [Rendering on the web](#rendering-on-the-web) (12)+- [Formats and asset pipeline](#formats-and-asset-pipeline) (12)+- [Physics, networking, and shared worlds](#physics-networking-and-shared-worlds) (9)+- [Agent frameworks and protocols](#agent-frameworks-and-protocols) (13)+- [Datasets and simulators](#datasets-and-simulators) (7)+- [Free assets](#free-assets) (6)+- [Learning](#learning) (7)++## Text and image to 3D++Generate a mesh from a prompt or a single photo. This is the front door of the pipeline: the fastest way to get geometry that did not exist an hour ago.++- [TRELLIS](https://github.com/microsoft/TRELLIS) - Structured latent generation that emits meshes, radiance fields, and Gaussians from the same model. Currently the strongest open text and image to 3D quality.+- [Hunyuan3D 2](https://github.com/Tencent-Hunyuan/Hunyuan3D-2) - Shape generation plus a separate texture synthesis stage, so the mesh and its material are not fighting for the same capacity. Ships usable PBR maps.+- [TripoSR](https://github.com/VAST-AI-Research/TripoSR) - Single image to 3D in well under a second on one GPU. The reference point for fast feed-forward reconstruction.+- [InstantMesh](https://github.com/TencentARC/InstantMesh) - Multi-view diffusion into a large reconstruction model, tuned for clean topology rather than the blobby output that plagues earlier feed-forward work.+- [Stable Fast 3D](https://github.com/Stability-AI/stable-fast-3d) - Image to UV-unwrapped, textured mesh in roughly half a second, with material parameter prediction built in.+- [LGM](https://github.com/3DTopia/LGM) - Large multi-view Gaussian model. High-resolution 3D from text or a single image, then converted to a mesh.+- [Wonder3D](https://github.com/xxlong0/Wonder3D) - Cross-domain diffusion that generates consistent normal maps and colour views together, which is why its surfaces hold up under relighting.+- [CRM](https://github.com/thu-ml/CRM) - Convolutional reconstruction model. Single image to a textured mesh in about ten seconds without per-object optimisation.+- [Zero-1-to-3](https://github.com/cvlab-columbia/zero123) - The novel-view-synthesis paper most of this section is built on. Learn what your favourite generator is actually doing by reading this one.+- [threestudio](https://github.com/threestudio-project/threestudio) - One framework, many text-to-3D methods (DreamFusion, Magic3D, ProlificDreamer, and more) behind a shared config system. The right place to compare approaches honestly.+- [Shap-E](https://github.com/openai/shap-e) - Generates implicit functions that render as both meshes and neural radiance fields. Old now, but small, permissive, and easy to run.+- [Point-E](https://github.com/openai/point-e) - Text to point cloud on a single GPU in a minute or two. Still the cheapest way to get rough 3D structure from a prompt.+- [three.ws Forge](https://three.ws/forge) - A free hosted text-to-3D lane with no key and no account. Returns a GLB you can download, embed, or rig.+- [Meshy](https://www.meshy.ai) - Commercial text and image to 3D with quad remeshing and PBR texturing. Good API, generous enough free tier to evaluate properly.+- [Tripo](https://www.tripo3d.ai) - The hosted product from the TripoSR authors. Fast previews, refined high-poly output, and a documented API.+- [Rodin](https://hyper3d.ai) - Generation aimed at production assets: clean topology options, PBR materials, and part-level control.++## Photos and video to 3D++Capture reality instead of inventing it. Photogrammetry, neural fields, and Gaussian splatting, plus the viewers that get the result onto the web.++- [3D Gaussian Splatting](https://github.com/graphdeco-inria/gaussian-splatting) - The original INRIA implementation that made radiance fields real-time. Everything in this section is downstream of it.+- [gsplat](https://github.com/nerfstudio-project/gsplat) - A CUDA-accelerated splatting library with a clean Python API, faster and far less memory-hungry than the reference code.+- [Nerfstudio](https://github.com/nerfstudio-project/nerfstudio) - A modular pipeline for neural radiance fields and splats, with a real-time web viewer and a sane data-processing story.+- [Spark](https://github.com/sparkjsdev/spark) - A Gaussian splatting renderer that drops into an existing three.js scene as ordinary objects, so splats and meshes light and sort together.+- [SuperSplat](https://github.com/playcanvas/supersplat) - A browser-based editor for cleaning, cropping, and compressing splat captures before you publish them.+- [Brush](https://github.com/ArthurBrussee/brush) - A splatting engine in Rust and wgpu that trains and renders on the web, macOS, Linux, Windows, and Android from one codebase.+- [OpenSplat](https://github.com/WebODM/OpenSplat) - Splat training on CPU as well as CUDA and Metal, which makes it the practical option on machines without an NVIDIA card.+- [COLMAP](https://github.com/colmap/colmap) - Structure from motion and multi-view stereo. The camera poses almost every method above expects as input come from here.+- [DUSt3R](https://github.com/naver/dust3r) - Dense 3D reconstruction from unconstrained image pairs with no camera calibration at all. It quietly deleted a whole preprocessing stage.+- [VGGT](https://github.com/facebookresearch/vggt) - A single feed-forward transformer that predicts camera poses, depth, and point maps together in seconds.+- [Meshroom](https://github.com/alicevision/Meshroom) - A node-graph photogrammetry application on top of AliceVision. The friendliest free path from a folder of photos to a textured mesh.+- [splat](https://github.com/antimatter15/splat) - A WebGL splat viewer in a few hundred readable lines. Read it once and you will understand the whole rendering technique.+- [Luma AI](https://lumalabs.ai) - Phone capture to a shareable splat, with an embed and a web SDK.+- [Polycam](https://poly.cam) - LiDAR and photo capture on phones with mesh, splat, and glTF export. The fastest route from a real object to a file you can use.++## Avatars and characters++Humanoid bodies that arrive already rigged, already skinned, and ready to be driven. Building one from scratch is rarely the right first move.++- [Avaturn](https://avaturn.me) - Photorealistic avatars from a phone photo, exported as GLB or FBX with a standard humanoid skeleton.+- [VRoid Studio](https://vroid.com/en/studio) - A free desktop editor for anime-style VRM characters, with hair, clothing, and expression authoring built in.+- [three-vrm](https://github.com/pixiv/three-vrm) - The VRM runtime for three.js: spring bones, look-at, expressions, and material handling that a plain glTF loader will not give you.+- [VRM specification](https://github.com/vrm-c/vrm-specification) - The glTF extension that standardises humanoid bones, expressions, look-at, and licence metadata. Read it before you invent your own avatar format.+- [MakeHuman](http://www.makehumancommunity.org) - An open parametric human generator with a full skeleton and topology you can actually edit. Old-fashioned, and still unmatched for free anatomically-driven bodies.+- [SMPL-X](https://smpl-x.is.tue.mpg.de) - A parametric body model with articulated hands and an expressive face. The shared coordinate system that most human motion research speaks.+- [Character Creator](https://www.reallusion.com/character-creator/) - Commercial character authoring with production-grade topology, morphs, and direct export to game engines.+- [three.ws Avatar Studio](https://three.ws/avatar-studio) - Generate, rig, dress, and pose an avatar in the browser, then embed it anywhere with one script tag.++## Rigging and retargeting++A mesh without a skeleton is a statue. This is the layer that decides whether an avatar can be animated at all.++- [Mixamo](https://www.mixamo.com) - Free automatic humanoid rigging plus a large motion-capture library. Its bone naming has become an informal industry convention.+- [UniRig](https://github.com/VAST-AI-Research/UniRig) - An autoregressive model that rigs humans, animals, and objects from one checkpoint, predicting the skeleton and the skin weights together.+- [RigNet](https://github.com/zhan-xu/RigNet) - The neural rigging paper this field grew out of: joint prediction, skeleton connectivity, and skinning learned end to end.+- [AccuRIG](https://actorcore.reallusion.com/auto-rig) - A free desktop auto-rigger for static humanoid meshes, with clean weight painting and direct export to standard skeletons.+- [Rigify](https://docs.blender.org/manual/en/4.5/addons/rigging/rigify/index.html) - Blender's bundled meta-rig system. Free, scriptable, and the practical fallback when an automatic rigger gets a character wrong.+- [SkeletonUtils](https://threejs.org/docs/#examples/en/utils/SkeletonUtils) - three.js retargeting helpers that map a clip authored for one skeleton onto another at runtime. Small, unglamorous, and load-bearing.++## Motion and animation++Generated motion, captured motion, and the datasets both are trained on. A body that only idles is a prop.++- [Motion Diffusion Model](https://github.com/GuyTevet/motion-diffusion-model) - Text to human motion via a lightweight diffusion model. The reference implementation nearly every later method compares against.+- [MoMask](https://github.com/EricGuo5513/momask-codes) - Generative masked modelling of motion tokens. Faster and sharper than diffusion baselines, with editing and in-betweening for free.+- [MotionGPT](https://github.com/OpenMotionLab/MotionGPT) - Treats motion as a language, so one model generates, captions, predicts, and in-betweens with the same interface.+- [T2M-GPT](https://github.com/Mael-zys/T2M-GPT) - VQ-VAE plus a transformer for text-driven motion. A clean, readable codebase to learn the tokenised-motion approach from.+- [OmniControl](https://github.com/neu-vi/OmniControl) - Adds spatial control signals to motion diffusion, so you can pin a hand to a location instead of hoping the prompt lands.+- [HumanML3D](https://github.com/EricGuo5513/HumanML3D) - The text-and-motion dataset behind most of this section. If you are benchmarking, you are benchmarking on this.+- [AMASS](https://amass.is.tue.mpg.de) - Fifteen-plus motion capture datasets unified onto one SMPL body. The reason cross-dataset motion training is possible at all.+- [Cascadeur](https://cascadeur.com) - Keyframe animation software with physics-aware assistance and AutoPosing, usable free for individuals under a revenue threshold.+- [Rokoko](https://www.rokoko.com) - Motion capture suits and a free video-to-motion tool, plus a Blender plugin that handles retargeting.++## Faces, lipsync, and voice++The difference between a model standing there and an agent talking to you. Speech in, speech out, and a mouth that agrees with the audio.++- [TalkingHead](https://github.com/met4citizen/TalkingHead) - A three.js class that turns a glTF humanoid avatar into a talking head with visemes, gaze, and mood. The most complete open reference for this exact problem.+- [Rhubarb Lip Sync](https://github.com/DanielSWolf/rhubarb-lip-sync) - Offline audio to mouth-shape timings as a plain command line tool. No model to host, no API to pay for.+- [MediaPipe](https://github.com/google-ai-edge/mediapipe) - Face landmarks and blendshape coefficients in the browser at video rate, which is enough to drive an avatar's face from a webcam.+- [ARKit blendshapes](https://developer.apple.com/documentation/arkit/arfaceanchor/blendshapelocation) - The 52-coefficient facial vocabulary that avatar formats, capture tools, and generators have all standardised on.+- [Whisper](https://github.com/openai/whisper) - Robust multilingual speech recognition that made self-hosted transcription the default rather than the exception.+- [faster-whisper](https://github.com/SYSTRAN/faster-whisper) - A CTranslate2 reimplementation of Whisper that is several times faster at the same accuracy. Use this in production, not the reference code.+- [whisper.cpp](https://github.com/ggml-org/whisper.cpp) - Whisper in plain C and C++ with no dependencies, running on phones, browsers via WebAssembly, and Raspberry Pis.+- [Kokoro](https://github.com/hexgrad/kokoro) - An 82M-parameter text to speech model with quality that embarrasses far larger ones, small enough to run in a browser tab.+- [Piper](https://github.com/rhasspy/piper) - Fast local neural text to speech built for low-power devices, with a large set of pretrained voices.+- [F5-TTS](https://github.com/SWivid/F5-TTS) - Flow-matching text to speech with convincing zero-shot voice cloning from a few seconds of reference audio.+- [Coqui TTS](https://github.com/coqui-ai/TTS) - A deep-learning toolkit covering dozens of text to speech and voice-conversion models, including XTTS multilingual cloning.+- [Silero VAD](https://github.com/snakers4/silero-vad) - Voice activity detection in under a millisecond per chunk. The unglamorous component that decides whether a voice agent feels responsive.+- [Pipecat](https://github.com/pipecat-ai/pipecat) - A Python framework for real-time voice and multimodal agents: pipelines, interruption handling, and transport out of the box.+- [LiveKit Agents](https://github.com/livekit/agents) - Build voice agents that join a WebRTC room, with turn detection and telephony handled for you.+- [Web Speech API](https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API) - Recognition and synthesis already in the browser, with zero bytes shipped. Always check this before adding a model.++## Rendering on the web++The runtime that actually puts the body on a page. Everything above is upstream of a choice made here.++- [three.js](https://github.com/mrdoob/three.js) - The default WebGL and WebGPU library for the web, with the deepest ecosystem of loaders, controls, and examples anywhere.+- [React Three Fiber](https://github.com/pmndrs/react-three-fiber) - three.js as a React renderer. Scene graphs become components, and the reconciler handles the mutation you would otherwise write by hand.+- [drei](https://github.com/pmndrs/drei) - The helper library that makes React Three Fiber practical: controls, loaders, staging, text, and effects you would otherwise rebuild every project.+- [Babylon.js](https://github.com/BabylonJS/Babylon.js) - A batteries-included web engine with a built-in physics layer, node material editor, and an inspector that is genuinely better than the alternatives.+- [PlayCanvas](https://github.com/playcanvas/engine) - A lightweight engine tuned hard for load time and mobile, backed by a collaborative browser editor.+- [model-viewer](https://github.com/google/model-viewer) - One custom element that renders a glTF with sensible defaults, accessibility, and AR on both phone platforms. The lowest-effort correct answer.+- [A-Frame](https://github.com/aframevr/aframe) - Declarative HTML for WebXR scenes, with an entity-component system underneath. Still the shortest path from markup to headset.+- [Threlte](https://github.com/threlte/threlte) - three.js for Svelte, with a component library, physics bindings, and a preprocessor that keeps bundles small.+- [TresJS](https://github.com/Tresjs/tres) - three.js as Vue components, with reactivity mapped onto scene-graph updates.+- [OGL](https://github.com/oframe/ogl) - A minimal WebGL library that stays close to the API instead of hiding it. The right tool when three.js is more engine than you need.+- [Needle Engine](https://needle.tools) - Author in Unity or Blender, ship a small web build. A pragmatic bridge for teams whose content pipeline already exists.+- [agent-3d](https://github.com/nirholas/three.ws/tree/main/avatar-sdk) - A web component that drops a rigged, animated, lip-syncing agent avatar onto any page in one tag, with a JS API for moods and gestures.++## Formats and asset pipeline++Getting the asset small, valid, and loadable. Most 3D-on-the-web failures are pipeline failures, not rendering failures.++- [glTF](https://github.com/KhronosGroup/glTF) - The transmission format for 3D on the web, plus its extension registry. If you are choosing a format, this is the answer.+- [glTF-Transform](https://github.com/donmccurdy/glTF-Transform) - Read, edit, optimise, and validate glTF from Node or the CLI. Deduplicate, resize textures, weld, and compress in a scripted pipeline.+- [meshoptimizer](https://github.com/zeux/meshoptimizer) - Vertex cache optimisation, simplification, and the meshopt compression that glTF files use. Its gltfpack CLI is the single highest-leverage step in most pipelines.+- [Draco](https://github.com/google/draco) - Geometry compression that shrinks meshes and point clouds hard, with decoders for the browser.+- [KTX-Software](https://github.com/KhronosGroup/KTX-Software) - Basis Universal texture transcoding. GPU-compressed textures that stay compressed in memory, which is usually the real budget problem.+- [glTF-Validator](https://github.com/KhronosGroup/glTF-Validator) - Tells you whether a file is actually valid before a user's browser does. Wire it into CI and stop guessing.+- [glTF Sample Assets](https://github.com/KhronosGroup/glTF-Sample-Assets) - The official test models, including deliberately awkward ones. Your loader is not done until these render correctly.+- [Blender](https://www.blender.org) - Free, scriptable in Python, and headless-capable, which makes it the workhorse of automated 3D pipelines as much as an authoring tool.+- [trimesh](https://github.com/mikedh/trimesh) - The Python library for loading, repairing, measuring, and booleaning meshes. Nearly every 3D backend service has this in it somewhere.+- [Assimp](https://github.com/assimp/assimp) - Imports roughly fifty 3D formats into one data structure. The escape hatch when someone sends you a file nothing else opens.+- [OpenUSD](https://github.com/PixarAnimationStudios/OpenUSD) - Pixar's scene description for large, layered, collaboratively edited worlds. Heavier than glTF, and the right answer above a certain scale.+- [three-mesh-bvh](https://github.com/gkjohnson/three-mesh-bvh) - Bounding volume hierarchies for three.js geometry, turning raycasts against dense meshes from unusable into instant.++## Physics, networking, and shared worlds++One avatar in a viewer is a demo. Several avatars in a place, with gravity and collisions, is a product.++- [Rapier](https://github.com/dimforge/rapier) - A Rust physics engine with deterministic cross-platform simulation and first-class WebAssembly bindings.+- [Jolt Physics](https://github.com/jrouwe/JoltPhysics) - The multi-core physics engine used in shipped AAA titles, with a WebAssembly build that runs in the browser.+- [cannon-es](https://github.com/pmndrs/cannon-es) - A maintained fork of cannon.js in modern JavaScript. Small and readable, and enough for character controllers and props.+- [Colyseus](https://github.com/colyseus/colyseus) - Authoritative multiplayer rooms with automatic state synchronisation and delta encoding, in Node.+- [Yjs](https://github.com/yjs/yjs) - CRDTs for shared editable state, with providers for WebRTC, WebSocket, and IndexedDB persistence.+- [LiveKit](https://github.com/livekit/livekit) - An open WebRTC server for real-time audio and video, with SDKs everywhere and a self-hostable path.+- [mediasoup](https://github.com/versatica/mediasoup) - A selective forwarding unit as a Node library rather than a service, for teams that want to own the media path.+- [geckos.io](https://github.com/geckosio/geckos.io) - Unreliable UDP-like messaging in the browser over WebRTC data channels, which is what fast-moving avatar positions actually want.+- [WebXR Device API](https://immersive-web.github.io/webxr/) - The specification behind every headset and phone AR session on the open web. Read the spec, not just a framework's wrapper.++## Agent frameworks and protocols++The mind that drives the body. Listed here because embodiment is a rendering problem bolted onto an agent problem, and most teams underestimate the second one.++- [Model Context Protocol](https://github.com/modelcontextprotocol) - The open protocol for giving a model tools, resources, and prompts from an external server. The cleanest way to let an agent act on a 3D pipeline.+- [Claude Agent SDK](https://github.com/anthropics/claude-agent-sdk-python) - Build agents on the same harness Claude Code runs on: tool loops, subagents, permissions, and session state.+- [OpenAI Agents SDK](https://github.com/openai/openai-agents-python) - A small framework for multi-agent handoffs, guardrails, and tracing, with a deliberately thin abstraction layer.+- [LangGraph](https://github.com/langchain-ai/langgraph) - Agents as explicit stateful graphs, with checkpointing and human-in-the-loop pauses. The right shape when a flow must be resumable.+- [Vercel AI SDK](https://github.com/vercel/ai) - One TypeScript interface across model providers, with streaming, structured output, and tool calls that work in the browser.+- [smolagents](https://github.com/huggingface/smolagents) - Agents that write Python to call their tools instead of emitting JSON, in about a thousand lines of library code.+- [CrewAI](https://github.com/crewAIInc/crewAI) - Role-based multi-agent orchestration with an emphasis on readable task delegation.+- [AutoGen](https://github.com/microsoft/autogen) - A research-grade framework for conversational multi-agent systems, with an event-driven core and a visual builder.+- [Mastra](https://github.com/mastra-ai/mastra) - A TypeScript agent framework with workflows, memory, and evals, aimed at shipping rather than research.+- [Letta](https://github.com/letta-ai/letta) - Agents with persistent memory as a first-class server concept, out of the MemGPT research line.+- [A2A](https://github.com/a2aproject/A2A) - An open protocol for agents built by different vendors to discover each other and collaborate over a common envelope.+- [Generative Agents](https://github.com/joonspk-research/generative_agents) - The Stanford simulation where twenty-five agents with memory, reflection, and planning produced emergent social behaviour. Still the clearest demonstration of why memory architecture matters.+- [Voyager](https://github.com/MineDojo/Voyager) - An embodied agent that plays Minecraft by writing and saving its own skills as code, then composing them. The best existing argument for embodiment as a learning signal.++## Datasets and simulators++What the models in this list were trained and evaluated on, and where embodied agents go to practise.++- [Objaverse](https://objaverse.allenai.org) - Around 800,000 annotated 3D objects, and the XL release pushes past ten million. The dataset that made open 3D generation viable.+- [Objaverse-XL](https://github.com/allenai/objaverse-xl) - The download and processing tooling for the ten-million-object release, including deduplication and rendering scripts.+- [ShapeNet](https://huggingface.co/datasets/ShapeNet/ShapeNetCore) - The categorised CAD-model dataset that anchored 3D deep learning for a decade, mirrored on Hugging Face while its original site is unreliable.+- [Habitat](https://github.com/facebookresearch/habitat-lab) - A fast photorealistic simulator for training embodied agents to navigate and manipulate in indoor scenes.+- [AI2-THOR](https://github.com/allenai/ai2thor) - Interactive 3D rooms where objects have state: doors open, eggs break, appliances turn on. Built for agents that change the world rather than tour it.+- [BEHAVIOR](https://behavior.stanford.edu) - A benchmark of a thousand everyday household activities defined in a logic language, with simulated physics and states.+- [ScanNet](http://www.scan-net.org) - Richly annotated RGB-D scans of real indoor spaces, and the benchmark most scene-understanding work still reports on.++## Free assets++Models, textures, and lighting you can ship without a lawyer. Every project in this list looks better with a real HDRI behind it.++- [Poly Haven](https://polyhaven.com) - HDRIs, PBR textures, and models, all CC0, all high quality, with no account required. Start here.+- [ambientCG](https://ambientcg.com) - Thousands of CC0 PBR material sets at multiple resolutions, downloadable in bulk.+- [Kenney](https://kenney.nl) - Tens of thousands of CC0 game assets, including clean low-poly 3D kits that prototype beautifully.+- [Quaternius](https://quaternius.com) - Stylised CC0 model packs with consistent proportions across sets, so a scene assembled from several packs still looks intentional.+- [Sketchfab](https://sketchfab.com) - The largest 3D model library on the web, with a downloadable Creative Commons subset and glTF export.+- [OpenGameArt](https://opengameart.org) - A long-running community archive of openly licensed art. Uneven, and it holds things that exist nowhere else.++## Learning++How to get good at this. Read in roughly this order.++- [three.js manual](https://threejs.org/manual/) - The official guide, rewritten around real problems rather than API surface. The fastest correct start.+- [Discover three.js](https://discoverthreejs.com) - A free book that teaches the concepts underneath the library, including colour management and performance work most tutorials skip.+- [Three.js Journey](https://threejs-journey.com) - Bruno Simon's paid course, and the most thorough path from nothing to shaders and production scenes.+- [The Book of Shaders](https://thebookofshaders.com) - An interactive introduction to fragment shaders. Work through it once and materials stop being magic.+- [WebGL Fundamentals](https://webglfundamentals.org) - What every abstraction in this list is hiding, explained from first principles with runnable examples.+- [WebGPU Fundamentals](https://webgpufundamentals.org) - The same treatment for WebGPU, which is where the web's rendering stack is heading.+- [Real-Time Rendering resources](https://www.realtimerendering.com) - The companion site to the standard reference text, with a maintained portal into graphics literature.++## Contributing++Additions are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md) first: entries live in [`data/awesome.json`](../data/awesome.json), not in this file, which is generated.++## License++The list is published under [Apache-2.0](../LICENSE) with the rest of this repository. Each linked project carries its own license.diff --git a/data/awesome.json b/data/awesome.jsonnew file mode 100644index 000000000..d992bc4f5--- /dev/null+++ b/data/awesome.json@@ -0,0 +1,1427 @@+{+ "$comment": "Source of truth for the Awesome 3D Agents list. Edit here, then run `npm run build:awesome` to regenerate awesome/README.md and public/awesome.json. Every url is checked by `npm run awesome:links`.",+ "slug": "awesome-3d-agents",+ "title": "Awesome 3D Agents",+ "tagline": "Everything you need to give an AI agent a body.",+ "intro": "A curated list of the models, engines, formats, and research that turn a language model into something you can see, animate, talk to, and put on a web page. Generation, rigging, motion, faces, voice, renderers, and the agent frameworks that drive them.",+ "repo": "https://github.com/nirholas/three.ws",+ "page": "https://three.ws/awesome",+ "sections": [+ {+ "id": "text-to-3d",+ "title": "Text and image to 3D",+ "description": "Generate a mesh from a prompt or a single photo. This is the front door of the pipeline: the fastest way to get geometry that did not exist an hour ago.",+ "items": [+ {+ "name": "TRELLIS",+ "url": "https://github.com/microsoft/TRELLIS",+ "description": "Structured latent generation that emits meshes, radiance fields, and Gaussians from the same model. Currently the strongest open text and image to 3D quality.",+ "tags": [+ "oss",+ "python",+ "image-to-3d"+ ]+ },+ {+ "name": "Hunyuan3D 2",+ "url": "https://github.com/Tencent-Hunyuan/Hunyuan3D-2",+ "description": "Shape generation plus a separate texture synthesis stage, so the mesh and its material are not fighting for the same capacity. Ships usable PBR maps.",+ "tags": [+ "oss",+ "python",+ "textures"+ ]+ },+ {+ "name": "TripoSR",+ "url": "https://github.com/VAST-AI-Research/TripoSR",+ "description": "Single image to 3D in well under a second on one GPU. The reference point for fast feed-forward reconstruction.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "InstantMesh",+ "url": "https://github.com/TencentARC/InstantMesh",+ "description": "Multi-view diffusion into a large reconstruction model, tuned for clean topology rather than the blobby output that plagues earlier feed-forward work.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Stable Fast 3D",+ "url": "https://github.com/Stability-AI/stable-fast-3d",+ "description": "Image to UV-unwrapped, textured mesh in roughly half a second, with material parameter prediction built in.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "LGM",+ "url": "https://github.com/3DTopia/LGM",+ "description": "Large multi-view Gaussian model. High-resolution 3D from text or a single image, then converted to a mesh.",+ "tags": [+ "oss",+ "python",+ "gaussians"+ ]+ },+ {+ "name": "Wonder3D",+ "url": "https://github.com/xxlong0/Wonder3D",+ "description": "Cross-domain diffusion that generates consistent normal maps and colour views together, which is why its surfaces hold up under relighting.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "CRM",+ "url": "https://github.com/thu-ml/CRM",+ "description": "Convolutional reconstruction model. Single image to a textured mesh in about ten seconds without per-object optimisation.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Zero-1-to-3",+ "url": "https://github.com/cvlab-columbia/zero123",+ "description": "The novel-view-synthesis paper most of this section is built on. Learn what your favourite generator is actually doing by reading this one.",+ "tags": [+ "oss",+ "research",+ "foundational"+ ]+ },+ {+ "name": "threestudio",+ "url": "https://github.com/threestudio-project/threestudio",+ "description": "One framework, many text-to-3D methods (DreamFusion, Magic3D, ProlificDreamer, and more) behind a shared config system. The right place to compare approaches honestly.",+ "tags": [+ "oss",+ "python",+ "framework"+ ]+ },+ {+ "name": "Shap-E",+ "url": "https://github.com/openai/shap-e",+ "description": "Generates implicit functions that render as both meshes and neural radiance fields. Old now, but small, permissive, and easy to run.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Point-E",+ "url": "https://github.com/openai/point-e",+ "description": "Text to point cloud on a single GPU in a minute or two. Still the cheapest way to get rough 3D structure from a prompt.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "three.ws Forge",+ "url": "https://three.ws/forge",+ "description": "A free hosted text-to-3D lane with no key and no account. Returns a GLB you can download, embed, or rig.",+ "tags": [+ "hosted",+ "free",+ "glb"+ ]+ },+ {+ "name": "Meshy",+ "url": "https://www.meshy.ai",+ "description": "Commercial text and image to 3D with quad remeshing and PBR texturing. Good API, generous enough free tier to evaluate properly.",+ "tags": [+ "service",+ "api"+ ]+ },+ {+ "name": "Tripo",+ "url": "https://www.tripo3d.ai",+ "description": "The hosted product from the TripoSR authors. Fast previews, refined high-poly output, and a documented API.",+ "tags": [+ "service",+ "api"+ ]+ },+ {+ "name": "Rodin",+ "url": "https://hyper3d.ai",+ "description": "Generation aimed at production assets: clean topology options, PBR materials, and part-level control.",+ "tags": [+ "service"+ ]+ }+ ]+ },+ {+ "id": "scan-to-3d",+ "title": "Photos and video to 3D",+ "description": "Capture reality instead of inventing it. Photogrammetry, neural fields, and Gaussian splatting, plus the viewers that get the result onto the web.",+ "items": [+ {+ "name": "3D Gaussian Splatting",+ "url": "https://github.com/graphdeco-inria/gaussian-splatting",+ "description": "The original INRIA implementation that made radiance fields real-time. Everything in this section is downstream of it.",+ "tags": [+ "oss",+ "research",+ "foundational"+ ]+ },+ {+ "name": "gsplat",+ "url": "https://github.com/nerfstudio-project/gsplat",+ "description": "A CUDA-accelerated splatting library with a clean Python API, faster and far less memory-hungry than the reference code.",+ "tags": [+ "oss",+ "python",+ "cuda"+ ]+ },+ {+ "name": "Nerfstudio",+ "url": "https://github.com/nerfstudio-project/nerfstudio",+ "description": "A modular pipeline for neural radiance fields and splats, with a real-time web viewer and a sane data-processing story.",+ "tags": [+ "oss",+ "python",+ "framework"+ ]+ },+ {+ "name": "Spark",+ "url": "https://github.com/sparkjsdev/spark",+ "description": "A Gaussian splatting renderer that drops into an existing three.js scene as ordinary objects, so splats and meshes light and sort together.",+ "tags": [+ "oss",+ "js",+ "threejs"+ ]+ },+ {+ "name": "SuperSplat",+ "url": "https://github.com/playcanvas/supersplat",+ "description": "A browser-based editor for cleaning, cropping, and compressing splat captures before you publish them.",+ "tags": [+ "oss",+ "js",+ "editor"+ ]+ },+ {+ "name": "Brush",+ "url": "https://github.com/ArthurBrussee/brush",+ "description": "A splatting engine in Rust and wgpu that trains and renders on the web, macOS, Linux, Windows, and Android from one codebase.",+ "tags": [+ "oss",+ "rust",+ "webgpu"+ ]+ },+ {+ "name": "OpenSplat",+ "url": "https://github.com/WebODM/OpenSplat",+ "description": "Splat training on CPU as well as CUDA and Metal, which makes it the practical option on machines without an NVIDIA card.",+ "tags": [+ "oss",+ "cpp"+ ]+ },+ {+ "name": "COLMAP",+ "url": "https://github.com/colmap/colmap",+ "description": "Structure from motion and multi-view stereo. The camera poses almost every method above expects as input come from here.",+ "tags": [+ "oss",+ "cpp",+ "foundational"+ ]+ },+ {+ "name": "DUSt3R",+ "url": "https://github.com/naver/dust3r",+ "description": "Dense 3D reconstruction from unconstrained image pairs with no camera calibration at all. It quietly deleted a whole preprocessing stage.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "VGGT",+ "url": "https://github.com/facebookresearch/vggt",+ "description": "A single feed-forward transformer that predicts camera poses, depth, and point maps together in seconds.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "Meshroom",+ "url": "https://github.com/alicevision/Meshroom",+ "description": "A node-graph photogrammetry application on top of AliceVision. The friendliest free path from a folder of photos to a textured mesh.",+ "tags": [+ "oss",+ "gui"+ ]+ },+ {+ "name": "splat",+ "url": "https://github.com/antimatter15/splat",+ "description": "A WebGL splat viewer in a few hundred readable lines. Read it once and you will understand the whole rendering technique.",+ "tags": [+ "oss",+ "js",+ "minimal"+ ]+ },+ {+ "name": "Luma AI",+ "url": "https://lumalabs.ai",+ "description": "Phone capture to a shareable splat, with an embed and a web SDK.",+ "tags": [+ "service",+ "mobile"+ ]+ },+ {+ "name": "Polycam",+ "url": "https://poly.cam",+ "description": "LiDAR and photo capture on phones with mesh, splat, and glTF export. The fastest route from a real object to a file you can use.",+ "tags": [+ "service",+ "mobile"+ ]+ }+ ]+ },+ {+ "id": "avatars",+ "title": "Avatars and characters",+ "description": "Humanoid bodies that arrive already rigged, already skinned, and ready to be driven. Building one from scratch is rarely the right first move.",+ "items": [+ {+ "name": "Avaturn",+ "url": "https://avaturn.me",+ "description": "Photorealistic avatars from a phone photo, exported as GLB or FBX with a standard humanoid skeleton.",+ "tags": [+ "service",+ "gltf"+ ]+ },+ {+ "name": "VRoid Studio",+ "url": "https://vroid.com/en/studio",+ "description": "A free desktop editor for anime-style VRM characters, with hair, clothing, and expression authoring built in.",+ "tags": [+ "free",+ "vrm",+ "editor"+ ]+ },+ {+ "name": "three-vrm",+ "url": "https://github.com/pixiv/three-vrm",+ "description": "The VRM runtime for three.js: spring bones, look-at, expressions, and material handling that a plain glTF loader will not give you.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "vrm"+ ]+ },+ {+ "name": "VRM specification",+ "url": "https://github.com/vrm-c/vrm-specification",+ "description": "The glTF extension that standardises humanoid bones, expressions, look-at, and licence metadata. Read it before you invent your own avatar format.",+ "tags": [+ "spec",+ "gltf"+ ]+ },+ {+ "name": "MakeHuman",+ "url": "http://www.makehumancommunity.org",+ "description": "An open parametric human generator with a full skeleton and topology you can actually edit. Old-fashioned, and still unmatched for free anatomically-driven bodies.",+ "tags": [+ "oss",+ "free"+ ]+ },+ {+ "name": "SMPL-X",+ "url": "https://smpl-x.is.tue.mpg.de",+ "description": "A parametric body model with articulated hands and an expressive face. The shared coordinate system that most human motion research speaks.",+ "tags": [+ "research",+ "model",+ "foundational"+ ]+ },+ {+ "name": "Character Creator",+ "url": "https://www.reallusion.com/character-creator/",+ "description": "Commercial character authoring with production-grade topology, morphs, and direct export to game engines.",+ "tags": [+ "service",+ "desktop"+ ]+ },+ {+ "name": "three.ws Avatar Studio",+ "url": "https://three.ws/avatar-studio",+ "description": "Generate, rig, dress, and pose an avatar in the browser, then embed it anywhere with one script tag.",+ "tags": [+ "hosted",+ "free",+ "gltf"+ ]+ }+ ]+ },+ {+ "id": "rigging",+ "title": "Rigging and retargeting",+ "description": "A mesh without a skeleton is a statue. This is the layer that decides whether an avatar can be animated at all.",+ "items": [+ {+ "name": "Mixamo",+ "url": "https://www.mixamo.com",+ "description": "Free automatic humanoid rigging plus a large motion-capture library. Its bone naming has become an informal industry convention.",+ "tags": [+ "free",+ "service",+ "animations"+ ]+ },+ {+ "name": "UniRig",+ "url": "https://github.com/VAST-AI-Research/UniRig",+ "description": "An autoregressive model that rigs humans, animals, and objects from one checkpoint, predicting the skeleton and the skin weights together.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "RigNet",+ "url": "https://github.com/zhan-xu/RigNet",+ "description": "The neural rigging paper this field grew out of: joint prediction, skeleton connectivity, and skinning learned end to end.",+ "tags": [+ "oss",+ "python",+ "research",+ "foundational"+ ]+ },+ {+ "name": "AccuRIG",+ "url": "https://actorcore.reallusion.com/auto-rig",+ "description": "A free desktop auto-rigger for static humanoid meshes, with clean weight painting and direct export to standard skeletons.",+ "tags": [+ "free",+ "desktop"+ ]+ },+ {+ "name": "Rigify",+ "url": "https://docs.blender.org/manual/en/4.5/addons/rigging/rigify/index.html",+ "description": "Blender's bundled meta-rig system. Free, scriptable, and the practical fallback when an automatic rigger gets a character wrong.",+ "tags": [+ "free",+ "blender"+ ]+ },+ {+ "name": "SkeletonUtils",+ "url": "https://threejs.org/docs/#examples/en/utils/SkeletonUtils",+ "description": "three.js retargeting helpers that map a clip authored for one skeleton onto another at runtime. Small, unglamorous, and load-bearing.",+ "tags": [+ "oss",+ "js",+ "threejs"+ ]+ }+ ]+ },+ {+ "id": "motion",+ "title": "Motion and animation",+ "description": "Generated motion, captured motion, and the datasets both are trained on. A body that only idles is a prop.",+ "items": [+ {+ "name": "Motion Diffusion Model",+ "url": "https://github.com/GuyTevet/motion-diffusion-model",+ "description": "Text to human motion via a lightweight diffusion model. The reference implementation nearly every later method compares against.",+ "tags": [+ "oss",+ "python",+ "research",+ "foundational"+ ]+ },+ {+ "name": "MoMask",+ "url": "https://github.com/EricGuo5513/momask-codes",+ "description": "Generative masked modelling of motion tokens. Faster and sharper than diffusion baselines, with editing and in-betweening for free.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "MotionGPT",+ "url": "https://github.com/OpenMotionLab/MotionGPT",+ "description": "Treats motion as a language, so one model generates, captions, predicts, and in-betweens with the same interface.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "T2M-GPT",+ "url": "https://github.com/Mael-zys/T2M-GPT",+ "description": "VQ-VAE plus a transformer for text-driven motion. A clean, readable codebase to learn the tokenised-motion approach from.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "OmniControl",+ "url": "https://github.com/neu-vi/OmniControl",+ "description": "Adds spatial control signals to motion diffusion, so you can pin a hand to a location instead of hoping the prompt lands.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "HumanML3D",+ "url": "https://github.com/EricGuo5513/HumanML3D",+ "description": "The text-and-motion dataset behind most of this section. If you are benchmarking, you are benchmarking on this.",+ "tags": [+ "dataset",+ "research"+ ]+ },+ {+ "name": "AMASS",+ "url": "https://amass.is.tue.mpg.de",+ "description": "Fifteen-plus motion capture datasets unified onto one SMPL body. The reason cross-dataset motion training is possible at all.",+ "tags": [+ "dataset",+ "research",+ "foundational"+ ]+ },+ {+ "name": "Cascadeur",+ "url": "https://cascadeur.com",+ "description": "Keyframe animation software with physics-aware assistance and AutoPosing, usable free for individuals under a revenue threshold.",+ "tags": [+ "desktop",+ "free-tier"+ ]+ },+ {+ "name": "Rokoko",+ "url": "https://www.rokoko.com",+ "description": "Motion capture suits and a free video-to-motion tool, plus a Blender plugin that handles retargeting.",+ "tags": [+ "service",+ "mocap"+ ]+ }+ ]+ },+ {+ "id": "face-voice",+ "title": "Faces, lipsync, and voice",+ "description": "The difference between a model standing there and an agent talking to you. Speech in, speech out, and a mouth that agrees with the audio.",+ "items": [+ {+ "name": "TalkingHead",+ "url": "https://github.com/met4citizen/TalkingHead",+ "description": "A three.js class that turns a glTF humanoid avatar into a talking head with visemes, gaze, and mood. The most complete open reference for this exact problem.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "lipsync"+ ]+ },+ {+ "name": "Rhubarb Lip Sync",+ "url": "https://github.com/DanielSWolf/rhubarb-lip-sync",+ "description": "Offline audio to mouth-shape timings as a plain command line tool. No model to host, no API to pay for.",+ "tags": [+ "oss",+ "cli",+ "lipsync"+ ]+ },+ {+ "name": "MediaPipe",+ "url": "https://github.com/google-ai-edge/mediapipe",+ "description": "Face landmarks and blendshape coefficients in the browser at video rate, which is enough to drive an avatar's face from a webcam.",+ "tags": [+ "oss",+ "js",+ "python",+ "tracking"+ ]+ },+ {+ "name": "ARKit blendshapes",+ "url": "https://developer.apple.com/documentation/arkit/arfaceanchor/blendshapelocation",+ "description": "The 52-coefficient facial vocabulary that avatar formats, capture tools, and generators have all standardised on.",+ "tags": [+ "spec",+ "reference"+ ]+ },+ {+ "name": "Whisper",+ "url": "https://github.com/openai/whisper",+ "description": "Robust multilingual speech recognition that made self-hosted transcription the default rather than the exception.",+ "tags": [+ "oss",+ "python",+ "stt",+ "foundational"+ ]+ },+ {+ "name": "faster-whisper",+ "url": "https://github.com/SYSTRAN/faster-whisper",+ "description": "A CTranslate2 reimplementation of Whisper that is several times faster at the same accuracy. Use this in production, not the reference code.",+ "tags": [+ "oss",+ "python",+ "stt"+ ]+ },+ {+ "name": "whisper.cpp",+ "url": "https://github.com/ggml-org/whisper.cpp",+ "description": "Whisper in plain C and C++ with no dependencies, running on phones, browsers via WebAssembly, and Raspberry Pis.",+ "tags": [+ "oss",+ "cpp",+ "stt",+ "edge"+ ]+ },+ {+ "name": "Kokoro",+ "url": "https://github.com/hexgrad/kokoro",+ "description": "An 82M-parameter text to speech model with quality that embarrasses far larger ones, small enough to run in a browser tab.",+ "tags": [+ "oss",+ "tts",+ "small"+ ]+ },+ {+ "name": "Piper",+ "url": "https://github.com/rhasspy/piper",+ "description": "Fast local neural text to speech built for low-power devices, with a large set of pretrained voices.",+ "tags": [+ "oss",+ "tts",+ "edge"+ ]+ },+ {+ "name": "F5-TTS",+ "url": "https://github.com/SWivid/F5-TTS",+ "description": "Flow-matching text to speech with convincing zero-shot voice cloning from a few seconds of reference audio.",+ "tags": [+ "oss",+ "tts",+ "cloning"+ ]+ },+ {+ "name": "Coqui TTS",+ "url": "https://github.com/coqui-ai/TTS",+ "description": "A deep-learning toolkit covering dozens of text to speech and voice-conversion models, including XTTS multilingual cloning.",+ "tags": [+ "oss",+ "python",+ "tts"+ ]+ },+ {+ "name": "Silero VAD",+ "url": "https://github.com/snakers4/silero-vad",+ "description": "Voice activity detection in under a millisecond per chunk. The unglamorous component that decides whether a voice agent feels responsive.",+ "tags": [+ "oss",+ "vad",+ "small"+ ]+ },+ {+ "name": "Pipecat",+ "url": "https://github.com/pipecat-ai/pipecat",+ "description": "A Python framework for real-time voice and multimodal agents: pipelines, interruption handling, and transport out of the box.",+ "tags": [+ "oss",+ "python",+ "realtime"+ ]+ },+ {+ "name": "LiveKit Agents",+ "url": "https://github.com/livekit/agents",+ "description": "Build voice agents that join a WebRTC room, with turn detection and telephony handled for you.",+ "tags": [+ "oss",+ "python",+ "realtime",+ "webrtc"+ ]+ },+ {+ "name": "Web Speech API",+ "url": "https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API",+ "description": "Recognition and synthesis already in the browser, with zero bytes shipped. Always check this before adding a model.",+ "tags": [+ "spec",+ "browser",+ "free"+ ]+ }+ ]+ },+ {+ "id": "renderers",+ "title": "Rendering on the web",+ "description": "The runtime that actually puts the body on a page. Everything above is upstream of a choice made here.",+ "items": [+ {+ "name": "three.js",+ "url": "https://github.com/mrdoob/three.js",+ "description": "The default WebGL and WebGPU library for the web, with the deepest ecosystem of loaders, controls, and examples anywhere.",+ "tags": [+ "oss",+ "js",+ "foundational"+ ]+ },+ {+ "name": "React Three Fiber",+ "url": "https://github.com/pmndrs/react-three-fiber",+ "description": "three.js as a React renderer. Scene graphs become components, and the reconciler handles the mutation you would otherwise write by hand.",+ "tags": [+ "oss",+ "js",+ "react",+ "threejs"+ ]+ },+ {+ "name": "drei",+ "url": "https://github.com/pmndrs/drei",+ "description": "The helper library that makes React Three Fiber practical: controls, loaders, staging, text, and effects you would otherwise rebuild every project.",+ "tags": [+ "oss",+ "js",+ "react"+ ]+ },+ {+ "name": "Babylon.js",+ "url": "https://github.com/BabylonJS/Babylon.js",+ "description": "A batteries-included web engine with a built-in physics layer, node material editor, and an inspector that is genuinely better than the alternatives.",+ "tags": [+ "oss",+ "js",+ "engine"+ ]+ },+ {+ "name": "PlayCanvas",+ "url": "https://github.com/playcanvas/engine",+ "description": "A lightweight engine tuned hard for load time and mobile, backed by a collaborative browser editor.",+ "tags": [+ "oss",+ "js",+ "engine"+ ]+ },+ {+ "name": "model-viewer",+ "url": "https://github.com/google/model-viewer",+ "description": "One custom element that renders a glTF with sensible defaults, accessibility, and AR on both phone platforms. The lowest-effort correct answer.",+ "tags": [+ "oss",+ "js",+ "webcomponent",+ "ar"+ ]+ },+ {+ "name": "A-Frame",+ "url": "https://github.com/aframevr/aframe",+ "description": "Declarative HTML for WebXR scenes, with an entity-component system underneath. Still the shortest path from markup to headset.",+ "tags": [+ "oss",+ "js",+ "webxr"+ ]+ },+ {+ "name": "Threlte",+ "url": "https://github.com/threlte/threlte",+ "description": "three.js for Svelte, with a component library, physics bindings, and a preprocessor that keeps bundles small.",+ "tags": [+ "oss",+ "js",+ "svelte"+ ]+ },+ {+ "name": "TresJS",+ "url": "https://github.com/Tresjs/tres",+ "description": "three.js as Vue components, with reactivity mapped onto scene-graph updates.",+ "tags": [+ "oss",+ "js",+ "vue"+ ]+ },+ {+ "name": "OGL",+ "url": "https://github.com/oframe/ogl",+ "description": "A minimal WebGL library that stays close to the API instead of hiding it. The right tool when three.js is more engine than you need.",+ "tags": [+ "oss",+ "js",+ "minimal"+ ]+ },+ {+ "name": "Needle Engine",+ "url": "https://needle.tools",+ "description": "Author in Unity or Blender, ship a small web build. A pragmatic bridge for teams whose content pipeline already exists.",+ "tags": [+ "service",+ "js",+ "pipeline"+ ]+ },+ {+ "name": "agent-3d",+ "url": "https://github.com/nirholas/three.ws/tree/main/avatar-sdk",+ "description": "A web component that drops a rigged, animated, lip-syncing agent avatar onto any page in one tag, with a JS API for moods and gestures.",+ "tags": [+ "oss",+ "js",+ "webcomponent",+ "avatars"+ ]+ }+ ]+ },+ {+ "id": "formats",+ "title": "Formats and asset pipeline",+ "description": "Getting the asset small, valid, and loadable. Most 3D-on-the-web failures are pipeline failures, not rendering failures.",+ "items": [+ {+ "name": "glTF",+ "url": "https://github.com/KhronosGroup/glTF",+ "description": "The transmission format for 3D on the web, plus its extension registry. If you are choosing a format, this is the answer.",+ "tags": [+ "spec",+ "foundational"+ ]+ },+ {+ "name": "glTF-Transform",+ "url": "https://github.com/donmccurdy/glTF-Transform",+ "description": "Read, edit, optimise, and validate glTF from Node or the CLI. Deduplicate, resize textures, weld, and compress in a scripted pipeline.",+ "tags": [+ "oss",+ "js",+ "cli",+ "pipeline"+ ]+ },+ {+ "name": "meshoptimizer",+ "url": "https://github.com/zeux/meshoptimizer",+ "description": "Vertex cache optimisation, simplification, and the meshopt compression that glTF files use. Its gltfpack CLI is the single highest-leverage step in most pipelines.",+ "tags": [+ "oss",+ "cpp",+ "compression",+ "foundational"+ ]+ },+ {+ "name": "Draco",+ "url": "https://github.com/google/draco",+ "description": "Geometry compression that shrinks meshes and point clouds hard, with decoders for the browser.",+ "tags": [+ "oss",+ "cpp",+ "compression"+ ]+ },+ {+ "name": "KTX-Software",+ "url": "https://github.com/KhronosGroup/KTX-Software",+ "description": "Basis Universal texture transcoding. GPU-compressed textures that stay compressed in memory, which is usually the real budget problem.",+ "tags": [+ "oss",+ "textures",+ "compression"+ ]+ },+ {+ "name": "glTF-Validator",+ "url": "https://github.com/KhronosGroup/glTF-Validator",+ "description": "Tells you whether a file is actually valid before a user's browser does. Wire it into CI and stop guessing.",+ "tags": [+ "oss",+ "validation",+ "ci"+ ]+ },+ {+ "name": "glTF Sample Assets",+ "url": "https://github.com/KhronosGroup/glTF-Sample-Assets",+ "description": "The official test models, including deliberately awkward ones. Your loader is not done until these render correctly.",+ "tags": [+ "assets",+ "testing"+ ]+ },+ {+ "name": "Blender",+ "url": "https://www.blender.org",+ "description": "Free, scriptable in Python, and headless-capable, which makes it the workhorse of automated 3D pipelines as much as an authoring tool.",+ "tags": [+ "oss",+ "free",+ "desktop",+ "python"+ ]+ },+ {+ "name": "trimesh",+ "url": "https://github.com/mikedh/trimesh",+ "description": "The Python library for loading, repairing, measuring, and booleaning meshes. Nearly every 3D backend service has this in it somewhere.",+ "tags": [+ "oss",+ "python",+ "foundational"+ ]+ },+ {+ "name": "Assimp",+ "url": "https://github.com/assimp/assimp",+ "description": "Imports roughly fifty 3D formats into one data structure. The escape hatch when someone sends you a file nothing else opens.",+ "tags": [+ "oss",+ "cpp",+ "conversion"+ ]+ },+ {+ "name": "OpenUSD",+ "url": "https://github.com/PixarAnimationStudios/OpenUSD",+ "description": "Pixar's scene description for large, layered, collaboratively edited worlds. Heavier than glTF, and the right answer above a certain scale.",+ "tags": [+ "oss",+ "cpp",+ "spec"+ ]+ },+ {+ "name": "three-mesh-bvh",+ "url": "https://github.com/gkjohnson/three-mesh-bvh",+ "description": "Bounding volume hierarchies for three.js geometry, turning raycasts against dense meshes from unusable into instant.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "performance"+ ]+ }+ ]+ },+ {+ "id": "worlds",+ "title": "Physics, networking, and shared worlds",+ "description": "One avatar in a viewer is a demo. Several avatars in a place, with gravity and collisions, is a product.",+ "items": [+ {+ "name": "Rapier",+ "url": "https://github.com/dimforge/rapier",+ "description": "A Rust physics engine with deterministic cross-platform simulation and first-class WebAssembly bindings.",+ "tags": [+ "oss",+ "rust",+ "wasm",+ "physics"+ ]+ },+ {+ "name": "Jolt Physics",+ "url": "https://github.com/jrouwe/JoltPhysics",+ "description": "The multi-core physics engine used in shipped AAA titles, with a WebAssembly build that runs in the browser.",+ "tags": [+ "oss",+ "cpp",+ "wasm",+ "physics"+ ]+ },+ {+ "name": "cannon-es",+ "url": "https://github.com/pmndrs/cannon-es",+ "description": "A maintained fork of cannon.js in modern JavaScript. Small and readable, and enough for character controllers and props.",+ "tags": [+ "oss",+ "js",+ "physics"+ ]+ },+ {+ "name": "Colyseus",+ "url": "https://github.com/colyseus/colyseus",+ "description": "Authoritative multiplayer rooms with automatic state synchronisation and delta encoding, in Node.",+ "tags": [+ "oss",+ "js",+ "multiplayer"+ ]+ },+ {+ "name": "Yjs",+ "url": "https://github.com/yjs/yjs",+ "description": "CRDTs for shared editable state, with providers for WebRTC, WebSocket, and IndexedDB persistence.",+ "tags": [+ "oss",+ "js",+ "crdt",+ "collaboration"+ ]+ },+ {+ "name": "LiveKit",+ "url": "https://github.com/livekit/livekit",+ "description": "An open WebRTC server for real-time audio and video, with SDKs everywhere and a self-hostable path.",+ "tags": [+ "oss",+ "go",+ "webrtc"+ ]+ },+ {+ "name": "mediasoup",+ "url": "https://github.com/versatica/mediasoup",+ "description": "A selective forwarding unit as a Node library rather than a service, for teams that want to own the media path.",+ "tags": [+ "oss",+ "js",+ "webrtc"+ ]+ },+ {+ "name": "geckos.io",+ "url": "https://github.com/geckosio/geckos.io",+ "description": "Unreliable UDP-like messaging in the browser over WebRTC data channels, which is what fast-moving avatar positions actually want.",+ "tags": [+ "oss",+ "js",+ "networking"+ ]+ },+ {+ "name": "WebXR Device API",+ "url": "https://immersive-web.github.io/webxr/",+ "description": "The specification behind every headset and phone AR session on the open web. Read the spec, not just a framework's wrapper.",+ "tags": [+ "spec",+ "webxr"+ ]+ }+ ]+ },+ {+ "id": "brains",+ "title": "Agent frameworks and protocols",+ "description": "The mind that drives the body. Listed here because embodiment is a rendering problem bolted onto an agent problem, and most teams underestimate the second one.",+ "items": [+ {+ "name": "Model Context Protocol",+ "url": "https://github.com/modelcontextprotocol",+ "description": "The open protocol for giving a model tools, resources, and prompts from an external server. The cleanest way to let an agent act on a 3D pipeline.",+ "tags": [+ "spec",+ "oss",+ "tools"+ ]+ },+ {+ "name": "Claude Agent SDK",+ "url": "https://github.com/anthropics/claude-agent-sdk-python",+ "description": "Build agents on the same harness Claude Code runs on: tool loops, subagents, permissions, and session state.",+ "tags": [+ "oss",+ "python",+ "sdk"+ ]+ },+ {+ "name": "OpenAI Agents SDK",+ "url": "https://github.com/openai/openai-agents-python",+ "description": "A small framework for multi-agent handoffs, guardrails, and tracing, with a deliberately thin abstraction layer.",+ "tags": [+ "oss",+ "python",+ "sdk"+ ]+ },+ {+ "name": "LangGraph",+ "url": "https://github.com/langchain-ai/langgraph",+ "description": "Agents as explicit stateful graphs, with checkpointing and human-in-the-loop pauses. The right shape when a flow must be resumable.",+ "tags": [+ "oss",+ "python",+ "js",+ "framework"+ ]+ },+ {+ "name": "Vercel AI SDK",+ "url": "https://github.com/vercel/ai",+ "description": "One TypeScript interface across model providers, with streaming, structured output, and tool calls that work in the browser.",+ "tags": [+ "oss",+ "js",+ "streaming"+ ]+ },+ {+ "name": "smolagents",+ "url": "https://github.com/huggingface/smolagents",+ "description": "Agents that write Python to call their tools instead of emitting JSON, in about a thousand lines of library code.",+ "tags": [+ "oss",+ "python",+ "minimal"+ ]+ },+ {+ "name": "CrewAI",+ "url": "https://github.com/crewAIInc/crewAI",+ "description": "Role-based multi-agent orchestration with an emphasis on readable task delegation.",+ "tags": [+ "oss",+ "python",+ "multi-agent"+ ]+ },+ {+ "name": "AutoGen",+ "url": "https://github.com/microsoft/autogen",+ "description": "A research-grade framework for conversational multi-agent systems, with an event-driven core and a visual builder.",+ "tags": [+ "oss",+ "python",+ "multi-agent"+ ]+ },+ {+ "name": "Mastra",+ "url": "https://github.com/mastra-ai/mastra",+ "description": "A TypeScript agent framework with workflows, memory, and evals, aimed at shipping rather than research.",+ "tags": [+ "oss",+ "js",+ "framework"+ ]+ },+ {+ "name": "Letta",+ "url": "https://github.com/letta-ai/letta",+ "description": "Agents with persistent memory as a first-class server concept, out of the MemGPT research line.",+ "tags": [+ "oss",+ "python",+ "memory"+ ]+ },+ {+ "name": "A2A",+ "url": "https://github.com/a2aproject/A2A",+ "description": "An open protocol for agents built by different vendors to discover each other and collaborate over a common envelope.",+ "tags": [+ "spec",+ "interop"+ ]+ },+ {+ "name": "Generative Agents",+ "url": "https://github.com/joonspk-research/generative_agents",+ "description": "The Stanford simulation where twenty-five agents with memory, reflection, and planning produced emergent social behaviour. Still the clearest demonstration of why memory architecture matters.",+ "tags": [+ "research",+ "foundational",+ "simulation"+ ]+ },+ {+ "name": "Voyager",+ "url": "https://github.com/MineDojo/Voyager",+ "description": "An embodied agent that plays Minecraft by writing and saving its own skills as code, then composing them. The best existing argument for embodiment as a learning signal.",+ "tags": [+ "research",+ "embodied",+ "foundational"+ ]+ }+ ]+ },+ {+ "id": "datasets",+ "title": "Datasets and simulators",+ "description": "What the models in this list were trained and evaluated on, and where embodied agents go to practise.",+ "items": [+ {+ "name": "Objaverse",+ "url": "https://objaverse.allenai.org",+ "description": "Around 800,000 annotated 3D objects, and the XL release pushes past ten million. The dataset that made open 3D generation viable.",+ "tags": [+ "dataset",+ "free",+ "foundational"+ ]+ },+ {+ "name": "Objaverse-XL",+ "url": "https://github.com/allenai/objaverse-xl",+ "description": "The download and processing tooling for the ten-million-object release, including deduplication and rendering scripts.",+ "tags": [+ "dataset",+ "python"+ ]+ },+ {+ "name": "ShapeNet",+ "url": "https://huggingface.co/datasets/ShapeNet/ShapeNetCore",+ "description": "The categorised CAD-model dataset that anchored 3D deep learning for a decade, mirrored on Hugging Face while its original site is unreliable.",+ "tags": [+ "dataset",+ "research"+ ]+ },+ {+ "name": "Habitat",+ "url": "https://github.com/facebookresearch/habitat-lab",+ "description": "A fast photorealistic simulator for training embodied agents to navigate and manipulate in indoor scenes.",+ "tags": [+ "oss",+ "python",+ "simulator"+ ]+ },+ {+ "name": "AI2-THOR",+ "url": "https://github.com/allenai/ai2thor",+ "description": "Interactive 3D rooms where objects have state: doors open, eggs break, appliances turn on. Built for agents that change the world rather than tour it.",+ "tags": [+ "oss",+ "python",+ "simulator"+ ]+ },+ {+ "name": "BEHAVIOR",+ "url": "https://behavior.stanford.edu",+ "description": "A benchmark of a thousand everyday household activities defined in a logic language, with simulated physics and states.",+ "tags": [+ "benchmark",+ "research"+ ]+ },+ {+ "name": "ScanNet",+ "url": "http://www.scan-net.org",+ "description": "Richly annotated RGB-D scans of real indoor spaces, and the benchmark most scene-understanding work still reports on.",+ "tags": [+ "dataset",+ "research"+ ]+ }+ ]+ },+ {+ "id": "assets",+ "title": "Free assets",+ "description": "Models, textures, and lighting you can ship without a lawyer. Every project in this list looks better with a real HDRI behind it.",+ "items": [+ {+ "name": "Poly Haven",+ "url": "https://polyhaven.com",+ "description": "HDRIs, PBR textures, and models, all CC0, all high quality, with no account required. Start here.",+ "tags": [+ "free",+ "cc0",+ "hdri",+ "textures"+ ]+ },+ {+ "name": "ambientCG",+ "url": "https://ambientcg.com",+ "description": "Thousands of CC0 PBR material sets at multiple resolutions, downloadable in bulk.",+ "tags": [+ "free",+ "cc0",+ "textures"+ ]+ },+ {+ "name": "Kenney",+ "url": "https://kenney.nl",+ "description": "Tens of thousands of CC0 game assets, including clean low-poly 3D kits that prototype beautifully.",+ "tags": [+ "free",+ "cc0",+ "models"+ ]+ },+ {+ "name": "Quaternius",+ "url": "https://quaternius.com",+ "description": "Stylised CC0 model packs with consistent proportions across sets, so a scene assembled from several packs still looks intentional.",+ "tags": [+ "free",+ "cc0",+ "models"+ ]+ },+ {+ "name": "Sketchfab",+ "url": "https://sketchfab.com",+ "description": "The largest 3D model library on the web, with a downloadable Creative Commons subset and glTF export.",+ "tags": [+ "service",+ "models"+ ]+ },+ {+ "name": "OpenGameArt",+ "url": "https://opengameart.org",+ "description": "A long-running community archive of openly licensed art. Uneven, and it holds things that exist nowhere else.",+ "tags": [+ "free",+ "models",+ "community"+ ]+ }+ ]+ },+ {+ "id": "learn",+ "title": "Learning",+ "description": "How to get good at this. Read in roughly this order.",+ "items": [+ {+ "name": "three.js manual",+ "url": "https://threejs.org/manual/",+ "description": "The official guide, rewritten around real problems rather than API surface. The fastest correct start.",+ "tags": [+ "free",+ "docs"+ ]+ },+ {+ "name": "Discover three.js",+ "url": "https://discoverthreejs.com",+ "description": "A free book that teaches the concepts underneath the library, including colour management and performance work most tutorials skip.",+ "tags": [+ "free",+ "book"+ ]+ },+ {+ "name": "Three.js Journey",+ "url": "https://threejs-journey.com",+ "description": "Bruno Simon's paid course, and the most thorough path from nothing to shaders and production scenes.",+ "tags": [+ "course",+ "paid"+ ]+ },+ {+ "name": "The Book of Shaders",+ "url": "https://thebookofshaders.com",+ "description": "An interactive introduction to fragment shaders. Work through it once and materials stop being magic.",+ "tags": [+ "free",+ "shaders",+ "interactive"+ ]+ },+ {+ "name": "WebGL Fundamentals",+ "url": "https://webglfundamentals.org",+ "description": "What every abstraction in this list is hiding, explained from first principles with runnable examples.",+ "tags": [+ "free",+ "webgl"+ ]+ },+ {+ "name": "WebGPU Fundamentals",+ "url": "https://webgpufundamentals.org",+ "description": "The same treatment for WebGPU, which is where the web's rendering stack is heading.",+ "tags": [+ "free",+ "webgpu"+ ]+ },+ {+ "name": "Real-Time Rendering resources",+ "url": "https://www.realtimerendering.com",+ "description": "The companion site to the standard reference text, with a maintained portal into graphics literature.",+ "tags": [+ "free",+ "reference",+ "graphics"+ ]+ }+ ]+ }+ ]+}diff --git a/data/changelog.json b/data/changelog.jsonindex 9429a1e6b..b5fcb7c5c 100644--- a/data/changelog.json+++ b/data/changelog.json@@ -1,6 +1,13 @@{"$comment": "Curated changelog entries for three.ws, the editorial layer of the public changelog. New PAGE launches are tracked automatically via the `added` field in data/pages.json; this file is for everything else holders care about: improvements, fixes, SDK releases, trading/payments capabilities, security work. scripts/build-page-index.mjs merges both sources into CHANGELOG.md, public/changelog.json, and public/changelog.xml. Newest first. Allowed tags: feature, improvement, fix, sdk, infra, docs, security. `link` is optional and must be a live three.ws path.","entries": [+ {+ "date": "2026-09-07",+ "title": "Awesome 3D Agents: a curated list of everything you need to give an AI a body",+ "summary": "Giving an agent a body means assembling a pipeline out of a dozen separate fields, and the map of that pipeline did not exist anywhere. There is now one: 134 hand-picked projects across 13 sections, covering text-to-3D generation, photo and video capture, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, physics and networking, agent frameworks, datasets, free assets, and where to learn all of it. Each entry says in one sentence what the thing does and why you would reach for it instead of the alternative. The list is searchable and filterable by tag at /awesome, copyable as Markdown, and published in the standard awesome-list format on GitHub so it can be read, forked, and contributed to there. Every link is fetched and verified before it ships.",+ "tags": ["feature", "docs"],+ "link": "/awesome"+ },{"date": "2026-09-07","title": "Tutorials and walkthroughs are now readable by search engines, and pages we never wanted indexed stopped asking to be",diff --git a/data/pages.json b/data/pages.jsonindex ea3ab9d32..2e0ea9350 100644--- a/data/pages.json+++ b/data/pages.json@@ -5818,6 +5818,15 @@"priority": 0.6,"changefreq": "monthly","added": "2026-09-05"+ },+ {+ "path": "/awesome",+ "title": "Awesome 3D Agents",+ "description": "A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.",+ "priority": 0.8,+ "changefreq": "weekly",+ "added": "2026-09-07",+ "showcase": true}]},diff --git a/package.json b/package.jsonindex 92b890d7e..dbc981272 100644--- a/package.json+++ b/package.json@@ -223,7 +223,7 @@"audit:route-shadowing": "node scripts/audit-route-shadowing.mjs","audit:handlers": "node scripts/audit-empty-handlers.mjs","audit:x402-catalog": "node scripts/audit-x402-catalog.mjs",- "gate": "npm run check:conflicts && npm run check:claude && npm run check:vendored && npm run check:model-viewer && npm run check:gcloudignore && npm run check:fetch-timeouts && npm run check:tdz-bootstrap && npm run check:event && npm run check:cron-syntax && npm run check:studio-openapi && npm run test:gate && npm run test:gate-3d && npm run audit:mcp && npm run audit:mcp-golden && npm run audit:mcp-safety && npm run audit:mcp-catalog && npm run audit:mcp-listing && npm run audit:routes && npm run audit:handlers && npm run audit:pages && npm run audit:links && npm run audit:docs && npm run audit:upstreams && npm run check:tutorials && npm run audit:hidden-guard && npm run audit:inline-handlers && npm run audit:x402-catalog && npm run audit:tokens && npm run audit:guards && npm run audit:tour-atlas && npm run check:ios-icons && npm run check:macos-icon && npm run check:apple-widget",+ "gate": "npm run check:conflicts && npm run check:claude && npm run check:vendored && npm run check:model-viewer && npm run check:gcloudignore && npm run check:fetch-timeouts && npm run check:tdz-bootstrap && npm run check:event && npm run check:cron-syntax && npm run check:studio-openapi && npm run test:gate && npm run test:gate-3d && npm run audit:mcp && npm run audit:mcp-golden && npm run audit:mcp-safety && npm run audit:mcp-catalog && npm run audit:mcp-listing && npm run audit:routes && npm run audit:handlers && npm run audit:pages && npm run audit:links && npm run check:awesome && npm run audit:docs && npm run audit:upstreams && npm run check:tutorials && npm run audit:hidden-guard && npm run audit:inline-handlers && npm run audit:x402-catalog && npm run audit:tokens && npm run audit:guards && npm run audit:tour-atlas && npm run check:ios-icons && npm run check:macos-icon && npm run check:apple-widget","test:gate-3d": "vitest run tests/glb-canonicalize.test.js tests/animation-retarget.test.js tests/animation-retarget-rest-pollution.test.js tests/viewer-framing.test.js","audit:mcp-golden": "node scripts/audit-mcp-golden.mjs","audit:mcp-safety": "node scripts/audit-mcp-safety.mjs",@@ -274,7 +274,7 @@"triage:gcp:deep": "node scripts/gcp-triage.mjs --deep --since 6h","snapshot": "node scripts/page-snapshot.mjs","build:news": "node scripts/build-news.mjs",- "prebuild": "node scripts/copy-three-decoders.mjs && node scripts/copy-voice-models.mjs && (node scripts/build-news.mjs & node scripts/build-skill-metadata.mjs & node scripts/build-local-skill-packs.mjs & node scripts/build-discovery-cards.mjs & node scripts/build-guards.mjs & node scripts/build-atlas-index.mjs & node scripts/build-mcp-catalog.mjs & npm run build:examples & npm run build:docs-search & npm run build:club-assets & npm run build:walk-environments & wait) && node scripts/inject-blog-seo.mjs --write && node scripts/inject-hidden-guard.mjs --write && (node scripts/build-page-index.mjs && node scripts/audit-page-index.mjs --strict && node scripts/audit-empty-handlers.mjs && node scripts/audit-hidden-guard.mjs) && node scripts/i18n-annotate.mjs --apply --pages=\"public/news/*.html\" && node scripts/inject-seo-meta.mjs --write && node scripts/inject-theme-boot.mjs --write",+ "prebuild": "node scripts/copy-three-decoders.mjs && node scripts/copy-voice-models.mjs && (node scripts/build-news.mjs & node scripts/build-skill-metadata.mjs & node scripts/build-local-skill-packs.mjs & node scripts/build-discovery-cards.mjs & node scripts/build-guards.mjs & node scripts/build-atlas-index.mjs & node scripts/build-mcp-catalog.mjs & node scripts/build-awesome.mjs & npm run build:examples & npm run build:docs-search & npm run build:club-assets & npm run build:walk-environments & wait) && node scripts/inject-blog-seo.mjs --write && node scripts/inject-hidden-guard.mjs --write && (node scripts/build-page-index.mjs && node scripts/audit-page-index.mjs --strict && node scripts/audit-empty-handlers.mjs && node scripts/audit-hidden-guard.mjs) && node scripts/i18n-annotate.mjs --apply --pages=\"public/news/*.html\" && node scripts/inject-seo-meta.mjs --write && node scripts/inject-theme-boot.mjs --write","seo:meta": "node scripts/inject-blog-seo.mjs --write && node scripts/build-page-index.mjs && node scripts/inject-seo-meta.mjs --write","build": "NODE_OPTIONS='--no-deprecation --max-old-space-size=6144' vite build && node scripts/strip-sw-from-embeds.mjs && node scripts/inject-tour-boot.mjs && node scripts/inject-atlas.mjs && node scripts/inject-inline-behaviors.mjs","build:wasm": "node scripts/build-wasm.mjs",@@ -454,7 +454,10 @@"test:home:live": "HOME_LIVE=1 vitest run packages/home-bridge tests/home-store.test.js","home:matrix": "node scripts/home-version-matrix.mjs","test:home:e2e": "node --env-file=.env.local ./node_modules/@playwright/test/cli.js test --config=playwright.home.config.js",- "deck:inception": "node scripts/build-inception-deck.mjs"+ "deck:inception": "node scripts/build-inception-deck.mjs",+ "build:awesome": "node scripts/build-awesome.mjs",+ "check:awesome": "node scripts/build-awesome.mjs --check",+ "awesome:links": "node scripts/check-awesome-links.mjs"},"dependencies": {"@0xsplits/splits-sdk": "^6.5.0",diff --git a/pages/awesome.html b/pages/awesome.htmlnew file mode 100644index 000000000..666e1910d--- /dev/null+++ b/pages/awesome.html@@ -0,0 +1,592 @@+<!DOCTYPE html>+<html lang="en">+<head>+ <script>/* three.ws theme boot: no-flash */(function(){try{var m=localStorage.getItem('twx_theme');var l=m==='auto'?(window.matchMedia&&window.matchMedia('(prefers-color-scheme: light)').matches):m==='light';document.documentElement.setAttribute('data-theme',l?'light':'dark');}catch(e){document.documentElement.setAttribute('data-theme','dark');}})();</script>+ <meta charset="utf-8" />+ <meta name="viewport" content="width=device-width, initial-scale=1, viewport-fit=cover" />+ <title>Awesome 3D Agents · three.ws</title>+ <meta name="description" content="A curated list of everything you need to give an AI agent a body: text-to-3D models, capture and splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable, filterable, and link-checked." />+ <link rel="canonical" href="https://three.ws/awesome" />+ <link rel="icon" href="/favicon.ico" sizes="any" />+ <link rel="icon" type="image/svg+xml" href="/favicon.svg" />+ <link rel="apple-touch-icon" href="/apple-touch-icon.png" />+ <link rel="manifest" href="/manifest.webmanifest" />+ <link rel="stylesheet" href="/style.css" />+ <link rel="stylesheet" href="/nav.css" />+ <link rel="stylesheet" href="/mobile.css" />+ <meta property="og:type" content="website" />+ <meta property="og:title" content="Awesome 3D Agents · three.ws" />+ <meta property="og:description" content="Everything you need to give an AI agent a body: generation, capture, avatars, rigging, motion, voice, renderers, pipelines, and agent frameworks. Curated and link-checked." />+ <meta property="og:url" content="https://three.ws/awesome" />+ <meta property="og:image" content="https://three.ws/api/page-og?s=learn&t=Awesome+3D+Agents&d=Everything+you+need+to+give+an+AI+agent+a+body%3A+generation%2C+capture%2C+avatars%2C+rigging%2C+motion%2C+voice%2C+renderers+and+the+frameworks+that+drive+them&p=%2Fawesome" />+ <meta name="twitter:card" content="summary_large_image" />+ <meta name="twitter:title" content="Awesome 3D Agents · three.ws" />+ <meta name="twitter:description" content="A curated, link-checked list of everything you need to give an AI agent a body." />+ <meta name="twitter:image" content="https://three.ws/api/page-og?s=learn&t=Awesome+3D+Agents&d=Everything+you+need+to+give+an+AI+agent+a+body%3A+generation%2C+capture%2C+avatars%2C+rigging%2C+motion%2C+voice%2C+renderers+and+the+frameworks+that+drive+them&p=%2Fawesome" />+ <meta name="twitter:site" content="@trythreews" />+ <meta name="robots" content="index, follow, max-image-preview:large, max-snippet:-1" />+ <meta property="og:locale" content="en_US" />+ <meta property="og:image:alt" content="Awesome 3D Agents / three.ws" />+ <style>+ :root {+ --aw-maxw: 1280px;+ --aw-rail: 216px;+ }+ * { box-sizing: border-box; }+ html, body { margin: 0; background: var(--bg-0); color: var(--ink); }+ body { font-family: var(--font-body, 'Inter', system-ui, sans-serif); min-height: 100vh; }+ .site-header { position: sticky; top: 0; z-index: 50; background: color-mix(in srgb, var(--bg-0) 82%, transparent); backdrop-filter: blur(10px); }+ :focus-visible { outline: 2px solid var(--accent); outline-offset: 2px; border-radius: 4px; }++ .aw-wrap { max-width: var(--aw-maxw); margin: 0 auto; padding: 0 18px 120px; }++ /* hero */+ .aw-hero { padding: 40px 0 6px; }+ .aw-eyebrow {+ display: inline-flex; align-items: center; gap: 8px; margin: 0 0 14px;+ padding: 5px 11px; border: 1px solid var(--stroke); border-radius: var(--radius-pill, 999px);+ font-size: 0.72rem; letter-spacing: 0.08em; text-transform: uppercase; color: var(--ink-dim);+ background: var(--surface-1, rgba(255,255,255,0.03));+ }+ .aw-eyebrow b { color: var(--ink-bright, var(--ink)); font-weight: 650; letter-spacing: 0.04em; }+ .aw-hero h1 {+ font-size: clamp(2rem, 6vw, 3.4rem); line-height: 1.05; margin: 0 0 12px;+ letter-spacing: -0.03em; color: var(--ink-bright, var(--ink));+ font-family: var(--font-display, inherit);+ }+ .aw-lede { margin: 0 0 10px; font-size: clamp(1.05rem, 2vw, 1.25rem); color: var(--ink-bright, var(--ink)); max-width: 46ch; line-height: 1.4; }+ .aw-intro { margin: 0; max-width: 70ch; color: var(--ink-dim); line-height: 1.65; }+ .aw-hero-actions { display: flex; flex-wrap: wrap; gap: 10px; margin-top: 22px; }++ .aw-btn {+ display: inline-flex; align-items: center; gap: 7px; padding: 9px 15px;+ font: inherit; font-size: 0.9rem; font-weight: 550; cursor: pointer;+ color: var(--ink); background: var(--surface-2, rgba(255,255,255,0.05));+ border: 1px solid var(--stroke); border-radius: var(--radius-md, 10px);+ text-decoration: none; transition: border-color 0.15s ease, background 0.15s ease, transform 0.1s ease;+ }+ .aw-btn:hover { border-color: var(--stroke-strong); background: var(--surface-3, rgba(255,255,255,0.08)); color: var(--ink-bright, var(--ink)); }+ .aw-btn:active { transform: translateY(1px); }+ .aw-btn-primary { background: var(--accent); border-color: transparent; color: var(--btn-primary-fg, #05070d); }+ .aw-btn-primary:hover { background: var(--btn-primary-bg-hover, var(--accent)); color: var(--btn-primary-fg, #05070d); filter: brightness(1.08); }+ .aw-btn svg { width: 15px; height: 15px; }++ /* stats */+ .aw-stats { display: grid; grid-template-columns: repeat(auto-fit, minmax(130px, 1fr)); gap: 10px; margin: 30px 0 4px; padding: 0; }+ .aw-stat { border: 1px solid var(--stroke); background: var(--surface-1, rgba(255,255,255,0.03)); border-radius: var(--radius-md, 10px); padding: 13px 15px; }+ .aw-stat dt { font-size: 0.7rem; text-transform: uppercase; letter-spacing: 0.07em; color: var(--ink-dim); margin: 0 0 5px; }+ .aw-stat dd { margin: 0; font-size: 1.5rem; font-weight: 650; font-variant-numeric: tabular-nums; color: var(--ink-bright, var(--ink)); line-height: 1; }++ /* controls */+ .aw-controls {+ position: sticky; top: 56px; z-index: 20; padding: 14px 0 12px; margin-bottom: 6px;+ background: color-mix(in srgb, var(--bg-0) 93%, transparent); backdrop-filter: blur(8px);+ border-bottom: 1px solid var(--hairline);+ }+ .aw-search { position: relative; display: flex; align-items: center; }+ .aw-search svg { position: absolute; left: 12px; width: 16px; height: 16px; color: var(--ink-dim); pointer-events: none; }+ .aw-search input {+ width: 100%; padding: 11px 60px 11px 36px; font: inherit; font-size: 0.95rem;+ color: var(--ink); background: var(--surface-2, rgba(255,255,255,0.05));+ border: 1px solid var(--stroke); border-radius: var(--radius-md, 10px);+ transition: border-color 0.15s ease, background 0.15s ease;+ }+ .aw-search input::placeholder { color: var(--ink-dim); }+ .aw-search input:hover { border-color: var(--stroke-strong); }+ .aw-search input:focus-visible { outline: none; border-color: var(--accent); background: var(--surface-3, rgba(255,255,255,0.07)); }+ .aw-kbd {+ position: absolute; right: 11px; pointer-events: none;+ padding: 2px 7px; font-size: 0.72rem; font-family: var(--font-mono, ui-monospace, monospace);+ color: var(--ink-dim); border: 1px solid var(--stroke); border-radius: 5px;+ background: var(--surface-1, rgba(255,255,255,0.04));+ }+ .aw-chips { display: flex; flex-wrap: wrap; align-items: center; gap: 7px; margin-top: 11px; }+ .aw-chips-label { font-size: 0.72rem; text-transform: uppercase; letter-spacing: 0.07em; color: var(--ink-dim); margin-right: 2px; }+ .aw-chip {+ padding: 5px 11px; font: inherit; font-size: 0.8rem; cursor: pointer;+ color: var(--ink-dim); background: transparent; border: 1px solid var(--stroke);+ border-radius: var(--radius-pill, 999px); transition: all 0.14s ease;+ }+ .aw-chip:hover { color: var(--ink-bright, var(--ink)); border-color: var(--stroke-strong); }+ .aw-chip[aria-pressed="true"] { color: var(--btn-primary-fg, #05070d); background: var(--accent); border-color: transparent; font-weight: 600; }+ .aw-chip small { opacity: 0.6; margin-left: 4px; font-variant-numeric: tabular-nums; }+ .aw-reset { margin-left: auto; }++ /* layout */+ .aw-layout { display: grid; grid-template-columns: 1fr; gap: 34px; margin-top: 26px; align-items: start; }+ @media (min-width: 1080px) { .aw-layout { grid-template-columns: var(--aw-rail) 1fr; } }++ .aw-rail { display: none; position: sticky; top: 132px; }+ @media (min-width: 1080px) { .aw-rail { display: block; } }+ .aw-rail h2 { font-size: 0.7rem; text-transform: uppercase; letter-spacing: 0.08em; color: var(--ink-dim); margin: 0 0 10px; }+ .aw-rail ol { list-style: none; margin: 0; padding: 0; display: flex; flex-direction: column; gap: 1px; }+ .aw-rail a {+ display: flex; justify-content: space-between; gap: 8px; padding: 6px 9px;+ font-size: 0.85rem; color: var(--ink-dim); text-decoration: none;+ border-radius: 7px; border-left: 2px solid transparent; transition: all 0.14s ease;+ }+ .aw-rail a:hover { color: var(--ink-bright, var(--ink)); background: var(--surface-1, rgba(255,255,255,0.04)); }+ .aw-rail a[aria-current="true"] { color: var(--ink-bright, var(--ink)); border-left-color: var(--accent); background: var(--surface-1, rgba(255,255,255,0.04)); }+ .aw-rail a span { font-variant-numeric: tabular-nums; opacity: 0.55; }++ /* sections */+ .aw-section { margin: 0 0 46px; scroll-margin-top: 140px; }+ .aw-section-head { margin-bottom: 16px; }+ .aw-section-head h2 { font-size: clamp(1.15rem, 2.4vw, 1.5rem); margin: 0 0 6px; letter-spacing: -0.015em; color: var(--ink-bright, var(--ink)); }+ .aw-anchor { color: var(--ink-dim); text-decoration: none; opacity: 0; margin-left: 8px; font-weight: 400; transition: opacity 0.14s ease; }+ .aw-section-head h2:hover .aw-anchor, .aw-anchor:focus-visible { opacity: 1; }+ .aw-section-head p { margin: 0; max-width: 74ch; color: var(--ink-dim); line-height: 1.6; font-size: 0.94rem; }++ .aw-grid { display: grid; grid-template-columns: repeat(auto-fill, minmax(268px, 1fr)); gap: 12px; }+ .aw-card {+ display: flex; flex-direction: column; gap: 8px; padding: 15px 16px;+ border: 1px solid var(--stroke); border-radius: var(--radius-md, 10px);+ background: var(--surface-1, rgba(255,255,255,0.025)); text-decoration: none; color: inherit;+ transition: border-color 0.16s ease, background 0.16s ease, transform 0.16s ease;+ }+ .aw-card:hover { border-color: var(--stroke-strong); background: var(--surface-2, rgba(255,255,255,0.05)); transform: translateY(-2px); }+ .aw-card:active { transform: translateY(0); }+ .aw-card-top { display: flex; align-items: baseline; justify-content: space-between; gap: 8px; }+ .aw-name { font-size: 0.98rem; font-weight: 620; color: var(--ink-bright, var(--ink)); letter-spacing: -0.01em; }+ .aw-host { font-size: 0.72rem; color: var(--ink-faint, var(--ink-dim)); font-family: var(--font-mono, ui-monospace, monospace); white-space: nowrap; }+ .aw-desc { margin: 0; font-size: 0.87rem; line-height: 1.55; color: var(--ink-dim); }+ .aw-tags { display: flex; flex-wrap: wrap; gap: 5px; margin-top: auto; padding-top: 4px; }+ .aw-tag {+ padding: 2px 7px; font-size: 0.68rem; letter-spacing: 0.02em;+ color: var(--ink-dim); border: 1px solid var(--hairline); border-radius: var(--radius-pill, 999px);+ }+ .aw-card mark { background: color-mix(in srgb, var(--accent) 30%, transparent); color: var(--ink-bright, var(--ink)); border-radius: 3px; padding: 0 1px; }++ /* states */+ .aw-count { margin: 0 0 18px; font-size: 0.85rem; color: var(--ink-dim); font-variant-numeric: tabular-nums; }+ .aw-skeleton { border: 1px solid var(--hairline); border-radius: var(--radius-md, 10px); padding: 15px 16px; display: flex; flex-direction: column; gap: 9px; }+ .aw-shimmer { height: 10px; border-radius: 5px; background: var(--skeleton-base, rgba(255,255,255,0.07)); animation: aw-pulse 1.4s ease-in-out infinite; }+ .aw-shimmer.w60 { width: 60%; } .aw-shimmer.w85 { width: 85%; } .aw-shimmer.w40 { width: 40%; }+ @keyframes aw-pulse { 0%, 100% { opacity: 0.45; } 50% { opacity: 0.9; } }+ @media (prefers-reduced-motion: reduce) {+ .aw-shimmer { animation: none; }+ .aw-card:hover { transform: none; }+ * { scroll-behavior: auto !important; }+ }+ .aw-empty, .aw-error {+ border: 1px dashed var(--stroke); border-radius: var(--radius-lg, 14px);+ padding: 34px 24px; text-align: center; background: var(--surface-1, rgba(255,255,255,0.02));+ }+ .aw-error { border-style: solid; border-color: color-mix(in srgb, var(--danger) 45%, var(--stroke)); }+ .aw-empty h2, .aw-error h2 { margin: 0 0 8px; font-size: 1.05rem; color: var(--ink-bright, var(--ink)); }+ .aw-empty p, .aw-error p { margin: 0 auto 16px; max-width: 52ch; color: var(--ink-dim); line-height: 1.6; font-size: 0.9rem; }++ /* contribute */+ .aw-cta {+ margin-top: 20px; padding: 26px 24px; border: 1px solid var(--stroke);+ border-radius: var(--radius-lg, 14px); background: var(--surface-1, rgba(255,255,255,0.03));+ }+ .aw-cta h2 { margin: 0 0 8px; font-size: 1.15rem; color: var(--ink-bright, var(--ink)); letter-spacing: -0.015em; }+ .aw-cta p { margin: 0 0 16px; max-width: 68ch; color: var(--ink-dim); line-height: 1.6; font-size: 0.92rem; }+ .aw-cta code { font-family: var(--font-mono, ui-monospace, monospace); font-size: 0.85em; padding: 1px 5px; border-radius: 4px; background: var(--surface-3, rgba(255,255,255,0.07)); }+ .aw-cta-actions { display: flex; flex-wrap: wrap; gap: 10px; }++ .aw-toast {+ position: fixed; left: 50%; bottom: 26px; transform: translate(-50%, 14px);+ padding: 10px 16px; border-radius: var(--radius-pill, 999px); font-size: 0.85rem;+ color: var(--ink-bright, var(--ink)); background: var(--surface-3, rgba(20,20,26,0.95));+ border: 1px solid var(--stroke-strong); box-shadow: var(--shadow-2, 0 8px 30px rgba(0,0,0,0.4));+ opacity: 0; pointer-events: none; transition: opacity 0.2s ease, transform 0.2s ease; z-index: 90;+ }+ .aw-toast[data-show="true"] { opacity: 1; transform: translate(-50%, 0); }+ </style>+</head>+<body>+ <header class="site-header"><div id="nav-container"></div></header>+ <script src="/nav.js"></script>++ <main class="aw-wrap">+ <section class="aw-hero">+ <p class="aw-eyebrow"><b>Awesome list</b> curated by three.ws</p>+ <h1 id="aw-title">Awesome 3D Agents</h1>+ <p class="aw-lede" id="aw-tagline">Everything you need to give an AI agent a body.</p>+ <p class="aw-intro" id="aw-intro">+ A curated list of the models, engines, formats, and research that turn a language model+ into something you can see, animate, talk to, and put on a web page.+ </p>+ <div class="aw-hero-actions">+ <a class="aw-btn aw-btn-primary" id="aw-repo" href="https://github.com/nirholas/three.ws/tree/main/awesome">+ <svg viewBox="0 0 16 16" fill="currentColor" aria-hidden="true"><path d="M8 0C3.58 0 0 3.58 0 8c0 3.54 2.29 6.53 5.47 7.59.4.07.55-.17.55-.38 0-.19-.01-.82-.01-1.49-2.01.37-2.53-.49-2.69-.94-.09-.23-.48-.94-.82-1.13-.28-.15-.68-.52-.01-.53.63-.01 1.08.58 1.23.82.72 1.21 1.87.87 2.33.66.07-.52.28-.87.51-1.07-1.78-.2-3.64-.89-3.64-3.95 0-.87.31-1.59.82-2.15-.08-.2-.36-1.02.08-2.12 0 0 .67-.21 2.2.82a7.4 7.4 0 0 1 2-.27c.68 0 1.36.09 2 .27 1.53-1.04 2.2-.82 2.2-.82.44 1.1.16 1.92.08 2.12.51.56.82 1.27.82 2.15 0 3.07-1.87 3.75-3.65 3.95.29.25.54.73.54 1.48 0 1.07-.01 1.93-.01 2.2 0 .21.15.46.55.38A8.01 8.01 0 0 0 16 8c0-4.42-3.58-8-8-8Z"/></svg>+ Star it on GitHub+ </a>+ <button type="button" class="aw-btn" id="aw-copy">+ <svg viewBox="0 0 16 16" fill="none" stroke="currentColor" stroke-width="1.5" aria-hidden="true"><rect x="5.5" y="5.5" width="8" height="9" rx="1.5"/><path d="M10.5 3.5h-7a1 1 0 0 0-1 1v8"/></svg>+ Copy as Markdown+ </button>+ <a class="aw-btn" id="aw-suggest" href="https://github.com/nirholas/three.ws/issues/new?title=Awesome%203D%20Agents%3A%20add%20%3Cproject%3E&body=Project%20url%3A%0ASection%3A%0AOne-sentence%20description%3A%0AWhy%20it%20belongs%3A">+ Suggest an entry+ </a>+ </div>+ <dl class="aw-stats" id="aw-stats" hidden></dl>+ </section>++ <div class="aw-controls">+ <div class="aw-search">+ <svg viewBox="0 0 16 16" fill="none" stroke="currentColor" stroke-width="1.6" aria-hidden="true"><circle cx="7" cy="7" r="4.5"/><path d="M10.5 10.5 14 14"/></svg>+ <label class="sr-only" for="aw-q">Search the list</label>+ <input id="aw-q" type="search" autocomplete="off" spellcheck="false" placeholder="Search the list…" />+ <span class="aw-kbd" aria-hidden="true">/</span>+ </div>+ <div class="aw-chips" id="aw-tagbar" role="group" aria-label="Filter by tag">+ <span class="aw-chips-label">Filter</span>+ </div>+ </div>++ <p class="aw-count" id="aw-count" role="status" aria-live="polite"></p>++ <div class="aw-layout">+ <nav class="aw-rail" id="aw-rail" aria-label="Sections">+ <h2>Sections</h2>+ <ol id="aw-rail-list"></ol>+ </nav>+ <div id="aw-sections"></div>+ </div>++ <section class="aw-cta">+ <h2>Something missing?</h2>+ <p>+ Entries live in <code>data/awesome.json</code>, and the list you are reading is generated+ from it, so a pull request never touches a Markdown file by hand. Every link here is+ fetched and classified by <code>npm run awesome:links</code> before it ships.+ </p>+ <div class="aw-cta-actions">+ <a class="aw-btn" href="https://github.com/nirholas/three.ws/blob/main/awesome/CONTRIBUTING.md">How to contribute</a>+ <a class="aw-btn" href="https://github.com/nirholas/three.ws/blob/main/data/awesome.json">Edit the source</a>+ <a class="aw-btn" href="/forge">Try the free text-to-3D lane</a>+ </div>+ </section>+ </main>++ <div class="aw-toast" id="aw-toast" role="status" aria-live="polite"></div>++ <script src="/brand.js" defer></script>+ <script type="module">+ const sectionsEl = document.getElementById('aw-sections');+ const railList = document.getElementById('aw-rail-list');+ const countEl = document.getElementById('aw-count');+ const statsEl = document.getElementById('aw-stats');+ const tagbar = document.getElementById('aw-tagbar');+ const qEl = document.getElementById('aw-q');+ const toastEl = document.getElementById('aw-toast');++ const state = { q: '', tags: new Set() };+ let data = null;++ const esc = (s) =>+ String(s ?? '').replace(/[&<>"']/g, (c) =>+ ({ '&': '&', '<': '<', '>': '>', '"': '"', "'": ''' })[c],+ );++ const hostOf = (url) => {+ try {+ return new URL(url).hostname.replace(/^www\./, '');+ } catch {+ return '';+ }+ };++ // Highlight the matched terms so a search result explains itself instead of+ // leaving the reader to scan the card for why it survived the filter.+ function highlight(text, terms) {+ const safe = esc(text);+ if (!terms.length) return safe;+ const pattern = terms+ .map((t) => t.replace(/[.*+?^${}()|[\]\\]/g, '\\$&'))+ .filter(Boolean)+ .join('|');+ if (!pattern) return safe;+ return safe.replace(new RegExp(`(${pattern})`, 'gi'), '<mark>$1</mark>');+ }++ let toastTimer;+ function toast(message) {+ toastEl.textContent = message;+ toastEl.dataset.show = 'true';+ clearTimeout(toastTimer);+ toastTimer = setTimeout(() => {+ toastEl.dataset.show = 'false';+ }, 2400);+ }++ function terms() {+ return state.q ? state.q.split(/\s+/).filter(Boolean) : [];+ }++ function matches(item, section) {+ for (const tag of state.tags) {+ if (!item.tags.includes(tag)) return false;+ }+ const list = terms();+ if (!list.length) return true;+ const hay = `${item.name} ${item.description} ${item.tags.join(' ')} ${section.title} ${hostOf(item.url)}`.toLowerCase();+ return list.every((t) => hay.includes(t));+ }++ function cardHtml(item, list) {+ return (+ `<a class="aw-card" href="${esc(item.url)}" target="_blank" rel="noopener">` ++ '<span class="aw-card-top">' ++ `<span class="aw-name">${highlight(item.name, list)}</span>` ++ `<span class="aw-host">${esc(hostOf(item.url))}</span>` ++ '</span>' ++ `<p class="aw-desc">${highlight(item.description, list)}</p>` ++ (item.tags.length+ ? `<span class="aw-tags">${item.tags.map((t) => `<span class="aw-tag">${esc(t)}</span>`).join('')}</span>`+ : '') ++ '</a>'+ );+ }++ function renderSkeleton() {+ countEl.textContent = 'Loading the list…';+ sectionsEl.innerHTML =+ '<div class="aw-grid">' ++ Array.from({ length: 9 })+ .map(+ () =>+ '<div class="aw-skeleton" aria-hidden="true"><div class="aw-shimmer w60"></div>' ++ '<div class="aw-shimmer w85"></div><div class="aw-shimmer"></div>' ++ '<div class="aw-shimmer w40"></div></div>',+ )+ .join('') ++ '</div>';+ }++ function renderError(message) {+ countEl.textContent = '';+ statsEl.hidden = true;+ sectionsEl.innerHTML =+ '<div class="aw-error"><h2>The list could not be loaded</h2>' ++ `<p>${esc(message)}</p>` ++ '<button type="button" class="aw-btn" id="aw-retry">Try again</button>' ++ '<p style="margin-top:14px">It is also readable on ' ++ '<a href="https://github.com/nirholas/three.ws/blob/main/awesome/README.md">GitHub</a>.</p></div>';+ document.getElementById('aw-retry').addEventListener('click', load);+ }++ function renderStats() {+ const cells = [+ ['Entries', data.counts.items, 'Curated projects, tools, specs, and papers'],+ ['Sections', data.counts.sections, 'One per stage of the pipeline'],+ ['Free to use', data.sections.flatMap((s) => s.items).filter((i) => i.tags.includes('oss') || i.tags.includes('free')).length, 'Open source or free to use'],+ ['Tags', data.counts.tags, 'Searchable facets across the list'],+ ];+ statsEl.innerHTML = cells+ .map(+ ([label, value, hint]) =>+ `<div class="aw-stat" title="${esc(hint)}"><dt>${esc(label)}</dt><dd>${value}</dd></div>`,+ )+ .join('');+ statsEl.hidden = false;+ }++ function renderTagbar() {+ const chips = data.tags+ .slice(0, 18)+ .map(+ (t) =>+ `<button type="button" class="aw-chip" data-tag="${esc(t.id)}" aria-pressed="false">` ++ `${esc(t.id)}<small>${t.count}</small></button>`,+ )+ .join('');+ tagbar.insertAdjacentHTML(+ 'beforeend',+ chips + '<button type="button" class="aw-btn aw-reset" id="aw-reset" hidden>Clear</button>',+ );+ tagbar.addEventListener('click', (event) => {+ const chip = event.target.closest('.aw-chip');+ if (chip) {+ const tag = chip.dataset.tag;+ if (state.tags.has(tag)) state.tags.delete(tag);+ else state.tags.add(tag);+ syncChips();+ syncUrl();+ render();+ return;+ }+ if (event.target.closest('#aw-reset')) clearFilters();+ });+ }++ function syncChips() {+ for (const chip of tagbar.querySelectorAll('.aw-chip')) {+ chip.setAttribute('aria-pressed', String(state.tags.has(chip.dataset.tag)));+ }+ const reset = document.getElementById('aw-reset');+ if (reset) reset.hidden = !state.q && !state.tags.size;+ }++ function clearFilters() {+ state.q = '';+ state.tags.clear();+ qEl.value = '';+ syncChips();+ syncUrl();+ render();+ qEl.focus();+ }++ function syncUrl() {+ const params = new URLSearchParams();+ if (state.q) params.set('q', state.q);+ if (state.tags.size) params.set('tag', [...state.tags].join(','));+ const qs = params.toString();+ history.replaceState(null, '', qs ? `?${qs}${location.hash}` : `${location.pathname}${location.hash}`);+ }++ function readUrl() {+ const params = new URLSearchParams(location.search);+ state.q = (params.get('q') ?? '').trim().toLowerCase();+ for (const t of (params.get('tag') ?? '').split(',').filter(Boolean)) state.tags.add(t);+ qEl.value = params.get('q') ?? '';+ }++ function render() {+ const list = terms();+ const visible = data.sections+ .map((section) => ({ section, items: section.items.filter((i) => matches(i, section)) }))+ .filter((group) => group.items.length);++ const shown = visible.reduce((n, g) => n + g.items.length, 0);+ const filtering = Boolean(state.q || state.tags.size);+ countEl.textContent = filtering+ ? `${shown} of ${data.counts.items} entries`+ : `${data.counts.items} entries across ${data.counts.sections} sections`;++ if (!shown) {+ sectionsEl.innerHTML =+ '<div class="aw-empty"><h2>Nothing matches those filters</h2>' ++ '<p>Every entry in the list is shown by default, so an empty result means the filters are ' ++ 'too narrow rather than the list being short. Try one term instead of two, or drop a tag.</p>' ++ '<button type="button" class="aw-btn" id="aw-clear">Clear filters</button></div>';+ document.getElementById('aw-clear').addEventListener('click', clearFilters);+ railList.innerHTML = '';+ return;+ }++ sectionsEl.innerHTML = visible+ .map(+ ({ section, items }) =>+ `<section class="aw-section" id="${esc(section.id)}">` ++ '<div class="aw-section-head">' ++ `<h2>${esc(section.title)}<a class="aw-anchor" href="#${esc(section.id)}" aria-label="Link to ${esc(section.title)}">#</a></h2>` ++ `<p>${esc(section.description)}</p></div>` ++ `<div class="aw-grid">${items.map((i) => cardHtml(i, list)).join('')}</div>` ++ '</section>',+ )+ .join('');++ railList.innerHTML = visible+ .map(+ ({ section, items }) =>+ `<li><a href="#${esc(section.id)}">${esc(section.title)}<span>${items.length}</span></a></li>`,+ )+ .join('');+ observeSections();+ }++ // Mark the section the reader is actually in, so the rail is a position+ // indicator rather than a plain table of contents.+ let observer = null;+ function observeSections() {+ observer?.disconnect();+ const links = new Map(+ [...railList.querySelectorAll('a')].map((a) => [a.getAttribute('href').slice(1), a]),+ );+ observer = new IntersectionObserver(+ (entries) => {+ for (const entry of entries) {+ if (!entry.isIntersecting) continue;+ for (const a of links.values()) a.removeAttribute('aria-current');+ links.get(entry.target.id)?.setAttribute('aria-current', 'true');+ }+ },+ { rootMargin: '-140px 0px -70% 0px', threshold: 0 },+ );+ for (const section of sectionsEl.querySelectorAll('.aw-section')) observer.observe(section);+ }++ function toMarkdown() {+ const lines = [`# ${data.title}`, '', `> ${data.tagline}`, '', data.intro, ''];+ for (const section of data.sections) {+ const items = section.items.filter((i) => matches(i, section));+ if (!items.length) continue;+ lines.push(`## ${section.title}`, '');+ for (const item of items) lines.push(`- [${item.name}](${item.url}) - ${item.description}`);+ lines.push('');+ }+ lines.push(`Source: ${location.origin}/awesome`);+ return lines.join('\n');+ }++ document.getElementById('aw-copy').addEventListener('click', async () => {+ try {+ await navigator.clipboard.writeText(toMarkdown());+ const filtering = Boolean(state.q || state.tags.size);+ toast(filtering ? 'Copied the filtered list as Markdown' : 'Copied the whole list as Markdown');+ } catch {+ toast('Your browser blocked the clipboard. Use the GitHub copy instead.');+ }+ });++ let debounce;+ qEl.addEventListener('input', () => {+ clearTimeout(debounce);+ debounce = setTimeout(() => {+ state.q = qEl.value.trim().toLowerCase();+ syncChips();+ syncUrl();+ render();+ }, 110);+ });++ document.addEventListener('keydown', (event) => {+ if (event.key === '/' && document.activeElement !== qEl && !event.metaKey && !event.ctrlKey) {+ event.preventDefault();+ qEl.focus();+ qEl.select();+ } else if (event.key === 'Escape' && document.activeElement === qEl) {+ clearFilters();+ }+ });++ async function load() {+ renderSkeleton();+ try {+ const res = await fetch('/awesome.json', {+ headers: { accept: 'application/json' },+ signal: AbortSignal.timeout(15000),+ });+ if (!res.ok) throw new Error(`the server returned ${res.status}`);+ data = await res.json();+ } catch (error) {+ renderError(+ `${error.message}. The list is a static file at /awesome.json, so this is usually a network problem.`,+ );+ return;+ }++ document.getElementById('aw-title').textContent = data.title;+ document.getElementById('aw-tagline').textContent = data.tagline;+ document.getElementById('aw-intro').textContent = data.intro;+ document.getElementById('aw-repo').href = data.repo;+ qEl.placeholder = `Search ${data.counts.items} entries by name, tag, or what they do…`;++ renderStats();+ renderTagbar();+ readUrl();+ syncChips();+ render();++ if (location.hash) document.getElementById(location.hash.slice(1))?.scrollIntoView();+ }++ load();+ </script>+</body>+</html>diff --git a/public/awesome.json b/public/awesome.jsonnew file mode 100644index 000000000..24e4f660c--- /dev/null+++ b/public/awesome.json@@ -0,0 +1,1616 @@+{+ "title": "Awesome 3D Agents",+ "tagline": "Everything you need to give an AI agent a body.",+ "intro": "A curated list of the models, engines, formats, and research that turn a language model into something you can see, animate, talk to, and put on a web page. Generation, rigging, motion, faces, voice, renderers, and the agent frameworks that drive them.",+ "repo": "https://github.com/nirholas/three.ws/tree/main/awesome",+ "markdown": "https://github.com/nirholas/three.ws/blob/main/awesome/README.md",+ "counts": {+ "items": 134,+ "sections": 13,+ "tags": 92+ },+ "tags": [+ {+ "id": "oss",+ "count": 88+ },+ {+ "id": "python",+ "count": 40+ },+ {+ "id": "js",+ "count": 29+ },+ {+ "id": "free",+ "count": 21+ },+ {+ "id": "research",+ "count": 20+ },+ {+ "id": "foundational",+ "count": 15+ },+ {+ "id": "service",+ "count": 11+ },+ {+ "id": "cpp",+ "count": 8+ },+ {+ "id": "spec",+ "count": 8+ },+ {+ "id": "dataset",+ "count": 6+ },+ {+ "id": "threejs",+ "count": 6+ },+ {+ "id": "cc0",+ "count": 4+ },+ {+ "id": "desktop",+ "count": 4+ },+ {+ "id": "framework",+ "count": 4+ },+ {+ "id": "models",+ "count": 4+ },+ {+ "id": "textures",+ "count": 4+ },+ {+ "id": "tts",+ "count": 4+ },+ {+ "id": "compression",+ "count": 3+ },+ {+ "id": "fast",+ "count": 3+ },+ {+ "id": "gltf",+ "count": 3+ },+ {+ "id": "minimal",+ "count": 3+ },+ {+ "id": "physics",+ "count": 3+ },+ {+ "id": "stt",+ "count": 3+ },+ {+ "id": "webrtc",+ "count": 3+ },+ {+ "id": "api",+ "count": 2+ },+ {+ "id": "cli",+ "count": 2+ },+ {+ "id": "edge",+ "count": 2+ },+ {+ "id": "editor",+ "count": 2+ },+ {+ "id": "engine",+ "count": 2+ },+ {+ "id": "hosted",+ "count": 2+ },+ {+ "id": "lipsync",+ "count": 2+ },+ {+ "id": "mobile",+ "count": 2+ },+ {+ "id": "multi-agent",+ "count": 2+ },+ {+ "id": "pipeline",+ "count": 2+ },+ {+ "id": "react",+ "count": 2+ },+ {+ "id": "realtime",+ "count": 2+ },+ {+ "id": "reference",+ "count": 2+ },+ {+ "id": "rust",+ "count": 2+ },+ {+ "id": "sdk",+ "count": 2+ },+ {+ "id": "simulator",+ "count": 2+ },+ {+ "id": "small",+ "count": 2+ },+ {+ "id": "vrm",+ "count": 2+ },+ {+ "id": "wasm",+ "count": 2+ },+ {+ "id": "webcomponent",+ "count": 2+ },+ {+ "id": "webgpu",+ "count": 2+ },+ {+ "id": "webxr",+ "count": 2+ }+ ],+ "sections": [+ {+ "id": "text-to-3d",+ "title": "Text and image to 3D",+ "description": "Generate a mesh from a prompt or a single photo. This is the front door of the pipeline: the fastest way to get geometry that did not exist an hour ago.",+ "items": [+ {+ "name": "TRELLIS",+ "url": "https://github.com/microsoft/TRELLIS",+ "description": "Structured latent generation that emits meshes, radiance fields, and Gaussians from the same model. Currently the strongest open text and image to 3D quality.",+ "tags": [+ "oss",+ "python",+ "image-to-3d"+ ]+ },+ {+ "name": "Hunyuan3D 2",+ "url": "https://github.com/Tencent-Hunyuan/Hunyuan3D-2",+ "description": "Shape generation plus a separate texture synthesis stage, so the mesh and its material are not fighting for the same capacity. Ships usable PBR maps.",+ "tags": [+ "oss",+ "python",+ "textures"+ ]+ },+ {+ "name": "TripoSR",+ "url": "https://github.com/VAST-AI-Research/TripoSR",+ "description": "Single image to 3D in well under a second on one GPU. The reference point for fast feed-forward reconstruction.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "InstantMesh",+ "url": "https://github.com/TencentARC/InstantMesh",+ "description": "Multi-view diffusion into a large reconstruction model, tuned for clean topology rather than the blobby output that plagues earlier feed-forward work.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Stable Fast 3D",+ "url": "https://github.com/Stability-AI/stable-fast-3d",+ "description": "Image to UV-unwrapped, textured mesh in roughly half a second, with material parameter prediction built in.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "LGM",+ "url": "https://github.com/3DTopia/LGM",+ "description": "Large multi-view Gaussian model. High-resolution 3D from text or a single image, then converted to a mesh.",+ "tags": [+ "oss",+ "python",+ "gaussians"+ ]+ },+ {+ "name": "Wonder3D",+ "url": "https://github.com/xxlong0/Wonder3D",+ "description": "Cross-domain diffusion that generates consistent normal maps and colour views together, which is why its surfaces hold up under relighting.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "CRM",+ "url": "https://github.com/thu-ml/CRM",+ "description": "Convolutional reconstruction model. Single image to a textured mesh in about ten seconds without per-object optimisation.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Zero-1-to-3",+ "url": "https://github.com/cvlab-columbia/zero123",+ "description": "The novel-view-synthesis paper most of this section is built on. Learn what your favourite generator is actually doing by reading this one.",+ "tags": [+ "oss",+ "research",+ "foundational"+ ]+ },+ {+ "name": "threestudio",+ "url": "https://github.com/threestudio-project/threestudio",+ "description": "One framework, many text-to-3D methods (DreamFusion, Magic3D, ProlificDreamer, and more) behind a shared config system. The right place to compare approaches honestly.",+ "tags": [+ "oss",+ "python",+ "framework"+ ]+ },+ {+ "name": "Shap-E",+ "url": "https://github.com/openai/shap-e",+ "description": "Generates implicit functions that render as both meshes and neural radiance fields. Old now, but small, permissive, and easy to run.",+ "tags": [+ "oss",+ "python"+ ]+ },+ {+ "name": "Point-E",+ "url": "https://github.com/openai/point-e",+ "description": "Text to point cloud on a single GPU in a minute or two. Still the cheapest way to get rough 3D structure from a prompt.",+ "tags": [+ "oss",+ "python",+ "fast"+ ]+ },+ {+ "name": "three.ws Forge",+ "url": "https://three.ws/forge",+ "description": "A free hosted text-to-3D lane with no key and no account. Returns a GLB you can download, embed, or rig.",+ "tags": [+ "hosted",+ "free",+ "glb"+ ]+ },+ {+ "name": "Meshy",+ "url": "https://www.meshy.ai",+ "description": "Commercial text and image to 3D with quad remeshing and PBR texturing. Good API, generous enough free tier to evaluate properly.",+ "tags": [+ "service",+ "api"+ ]+ },+ {+ "name": "Tripo",+ "url": "https://www.tripo3d.ai",+ "description": "The hosted product from the TripoSR authors. Fast previews, refined high-poly output, and a documented API.",+ "tags": [+ "service",+ "api"+ ]+ },+ {+ "name": "Rodin",+ "url": "https://hyper3d.ai",+ "description": "Generation aimed at production assets: clean topology options, PBR materials, and part-level control.",+ "tags": [+ "service"+ ]+ }+ ]+ },+ {+ "id": "scan-to-3d",+ "title": "Photos and video to 3D",+ "description": "Capture reality instead of inventing it. Photogrammetry, neural fields, and Gaussian splatting, plus the viewers that get the result onto the web.",+ "items": [+ {+ "name": "3D Gaussian Splatting",+ "url": "https://github.com/graphdeco-inria/gaussian-splatting",+ "description": "The original INRIA implementation that made radiance fields real-time. Everything in this section is downstream of it.",+ "tags": [+ "oss",+ "research",+ "foundational"+ ]+ },+ {+ "name": "gsplat",+ "url": "https://github.com/nerfstudio-project/gsplat",+ "description": "A CUDA-accelerated splatting library with a clean Python API, faster and far less memory-hungry than the reference code.",+ "tags": [+ "oss",+ "python",+ "cuda"+ ]+ },+ {+ "name": "Nerfstudio",+ "url": "https://github.com/nerfstudio-project/nerfstudio",+ "description": "A modular pipeline for neural radiance fields and splats, with a real-time web viewer and a sane data-processing story.",+ "tags": [+ "oss",+ "python",+ "framework"+ ]+ },+ {+ "name": "Spark",+ "url": "https://github.com/sparkjsdev/spark",+ "description": "A Gaussian splatting renderer that drops into an existing three.js scene as ordinary objects, so splats and meshes light and sort together.",+ "tags": [+ "oss",+ "js",+ "threejs"+ ]+ },+ {+ "name": "SuperSplat",+ "url": "https://github.com/playcanvas/supersplat",+ "description": "A browser-based editor for cleaning, cropping, and compressing splat captures before you publish them.",+ "tags": [+ "oss",+ "js",+ "editor"+ ]+ },+ {+ "name": "Brush",+ "url": "https://github.com/ArthurBrussee/brush",+ "description": "A splatting engine in Rust and wgpu that trains and renders on the web, macOS, Linux, Windows, and Android from one codebase.",+ "tags": [+ "oss",+ "rust",+ "webgpu"+ ]+ },+ {+ "name": "OpenSplat",+ "url": "https://github.com/WebODM/OpenSplat",+ "description": "Splat training on CPU as well as CUDA and Metal, which makes it the practical option on machines without an NVIDIA card.",+ "tags": [+ "oss",+ "cpp"+ ]+ },+ {+ "name": "COLMAP",+ "url": "https://github.com/colmap/colmap",+ "description": "Structure from motion and multi-view stereo. The camera poses almost every method above expects as input come from here.",+ "tags": [+ "oss",+ "cpp",+ "foundational"+ ]+ },+ {+ "name": "DUSt3R",+ "url": "https://github.com/naver/dust3r",+ "description": "Dense 3D reconstruction from unconstrained image pairs with no camera calibration at all. It quietly deleted a whole preprocessing stage.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "VGGT",+ "url": "https://github.com/facebookresearch/vggt",+ "description": "A single feed-forward transformer that predicts camera poses, depth, and point maps together in seconds.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "Meshroom",+ "url": "https://github.com/alicevision/Meshroom",+ "description": "A node-graph photogrammetry application on top of AliceVision. The friendliest free path from a folder of photos to a textured mesh.",+ "tags": [+ "oss",+ "gui"+ ]+ },+ {+ "name": "splat",+ "url": "https://github.com/antimatter15/splat",+ "description": "A WebGL splat viewer in a few hundred readable lines. Read it once and you will understand the whole rendering technique.",+ "tags": [+ "oss",+ "js",+ "minimal"+ ]+ },+ {+ "name": "Luma AI",+ "url": "https://lumalabs.ai",+ "description": "Phone capture to a shareable splat, with an embed and a web SDK.",+ "tags": [+ "service",+ "mobile"+ ]+ },+ {+ "name": "Polycam",+ "url": "https://poly.cam",+ "description": "LiDAR and photo capture on phones with mesh, splat, and glTF export. The fastest route from a real object to a file you can use.",+ "tags": [+ "service",+ "mobile"+ ]+ }+ ]+ },+ {+ "id": "avatars",+ "title": "Avatars and characters",+ "description": "Humanoid bodies that arrive already rigged, already skinned, and ready to be driven. Building one from scratch is rarely the right first move.",+ "items": [+ {+ "name": "Avaturn",+ "url": "https://avaturn.me",+ "description": "Photorealistic avatars from a phone photo, exported as GLB or FBX with a standard humanoid skeleton.",+ "tags": [+ "service",+ "gltf"+ ]+ },+ {+ "name": "VRoid Studio",+ "url": "https://vroid.com/en/studio",+ "description": "A free desktop editor for anime-style VRM characters, with hair, clothing, and expression authoring built in.",+ "tags": [+ "free",+ "vrm",+ "editor"+ ]+ },+ {+ "name": "three-vrm",+ "url": "https://github.com/pixiv/three-vrm",+ "description": "The VRM runtime for three.js: spring bones, look-at, expressions, and material handling that a plain glTF loader will not give you.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "vrm"+ ]+ },+ {+ "name": "VRM specification",+ "url": "https://github.com/vrm-c/vrm-specification",+ "description": "The glTF extension that standardises humanoid bones, expressions, look-at, and licence metadata. Read it before you invent your own avatar format.",+ "tags": [+ "spec",+ "gltf"+ ]+ },+ {+ "name": "MakeHuman",+ "url": "http://www.makehumancommunity.org",+ "description": "An open parametric human generator with a full skeleton and topology you can actually edit. Old-fashioned, and still unmatched for free anatomically-driven bodies.",+ "tags": [+ "oss",+ "free"+ ]+ },+ {+ "name": "SMPL-X",+ "url": "https://smpl-x.is.tue.mpg.de",+ "description": "A parametric body model with articulated hands and an expressive face. The shared coordinate system that most human motion research speaks.",+ "tags": [+ "research",+ "model",+ "foundational"+ ]+ },+ {+ "name": "Character Creator",+ "url": "https://www.reallusion.com/character-creator/",+ "description": "Commercial character authoring with production-grade topology, morphs, and direct export to game engines.",+ "tags": [+ "service",+ "desktop"+ ]+ },+ {+ "name": "three.ws Avatar Studio",+ "url": "https://three.ws/avatar-studio",+ "description": "Generate, rig, dress, and pose an avatar in the browser, then embed it anywhere with one script tag.",+ "tags": [+ "hosted",+ "free",+ "gltf"+ ]+ }+ ]+ },+ {+ "id": "rigging",+ "title": "Rigging and retargeting",+ "description": "A mesh without a skeleton is a statue. This is the layer that decides whether an avatar can be animated at all.",+ "items": [+ {+ "name": "Mixamo",+ "url": "https://www.mixamo.com",+ "description": "Free automatic humanoid rigging plus a large motion-capture library. Its bone naming has become an informal industry convention.",+ "tags": [+ "free",+ "service",+ "animations"+ ]+ },+ {+ "name": "UniRig",+ "url": "https://github.com/VAST-AI-Research/UniRig",+ "description": "An autoregressive model that rigs humans, animals, and objects from one checkpoint, predicting the skeleton and the skin weights together.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "RigNet",+ "url": "https://github.com/zhan-xu/RigNet",+ "description": "The neural rigging paper this field grew out of: joint prediction, skeleton connectivity, and skinning learned end to end.",+ "tags": [+ "oss",+ "python",+ "research",+ "foundational"+ ]+ },+ {+ "name": "AccuRIG",+ "url": "https://actorcore.reallusion.com/auto-rig",+ "description": "A free desktop auto-rigger for static humanoid meshes, with clean weight painting and direct export to standard skeletons.",+ "tags": [+ "free",+ "desktop"+ ]+ },+ {+ "name": "Rigify",+ "url": "https://docs.blender.org/manual/en/4.5/addons/rigging/rigify/index.html",+ "description": "Blender's bundled meta-rig system. Free, scriptable, and the practical fallback when an automatic rigger gets a character wrong.",+ "tags": [+ "free",+ "blender"+ ]+ },+ {+ "name": "SkeletonUtils",+ "url": "https://threejs.org/docs/#examples/en/utils/SkeletonUtils",+ "description": "three.js retargeting helpers that map a clip authored for one skeleton onto another at runtime. Small, unglamorous, and load-bearing.",+ "tags": [+ "oss",+ "js",+ "threejs"+ ]+ }+ ]+ },+ {+ "id": "motion",+ "title": "Motion and animation",+ "description": "Generated motion, captured motion, and the datasets both are trained on. A body that only idles is a prop.",+ "items": [+ {+ "name": "Motion Diffusion Model",+ "url": "https://github.com/GuyTevet/motion-diffusion-model",+ "description": "Text to human motion via a lightweight diffusion model. The reference implementation nearly every later method compares against.",+ "tags": [+ "oss",+ "python",+ "research",+ "foundational"+ ]+ },+ {+ "name": "MoMask",+ "url": "https://github.com/EricGuo5513/momask-codes",+ "description": "Generative masked modelling of motion tokens. Faster and sharper than diffusion baselines, with editing and in-betweening for free.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "MotionGPT",+ "url": "https://github.com/OpenMotionLab/MotionGPT",+ "description": "Treats motion as a language, so one model generates, captions, predicts, and in-betweens with the same interface.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "T2M-GPT",+ "url": "https://github.com/Mael-zys/T2M-GPT",+ "description": "VQ-VAE plus a transformer for text-driven motion. A clean, readable codebase to learn the tokenised-motion approach from.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "OmniControl",+ "url": "https://github.com/neu-vi/OmniControl",+ "description": "Adds spatial control signals to motion diffusion, so you can pin a hand to a location instead of hoping the prompt lands.",+ "tags": [+ "oss",+ "python",+ "research"+ ]+ },+ {+ "name": "HumanML3D",+ "url": "https://github.com/EricGuo5513/HumanML3D",+ "description": "The text-and-motion dataset behind most of this section. If you are benchmarking, you are benchmarking on this.",+ "tags": [+ "dataset",+ "research"+ ]+ },+ {+ "name": "AMASS",+ "url": "https://amass.is.tue.mpg.de",+ "description": "Fifteen-plus motion capture datasets unified onto one SMPL body. The reason cross-dataset motion training is possible at all.",+ "tags": [+ "dataset",+ "research",+ "foundational"+ ]+ },+ {+ "name": "Cascadeur",+ "url": "https://cascadeur.com",+ "description": "Keyframe animation software with physics-aware assistance and AutoPosing, usable free for individuals under a revenue threshold.",+ "tags": [+ "desktop",+ "free-tier"+ ]+ },+ {+ "name": "Rokoko",+ "url": "https://www.rokoko.com",+ "description": "Motion capture suits and a free video-to-motion tool, plus a Blender plugin that handles retargeting.",+ "tags": [+ "service",+ "mocap"+ ]+ }+ ]+ },+ {+ "id": "face-voice",+ "title": "Faces, lipsync, and voice",+ "description": "The difference between a model standing there and an agent talking to you. Speech in, speech out, and a mouth that agrees with the audio.",+ "items": [+ {+ "name": "TalkingHead",+ "url": "https://github.com/met4citizen/TalkingHead",+ "description": "A three.js class that turns a glTF humanoid avatar into a talking head with visemes, gaze, and mood. The most complete open reference for this exact problem.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "lipsync"+ ]+ },+ {+ "name": "Rhubarb Lip Sync",+ "url": "https://github.com/DanielSWolf/rhubarb-lip-sync",+ "description": "Offline audio to mouth-shape timings as a plain command line tool. No model to host, no API to pay for.",+ "tags": [+ "oss",+ "cli",+ "lipsync"+ ]+ },+ {+ "name": "MediaPipe",+ "url": "https://github.com/google-ai-edge/mediapipe",+ "description": "Face landmarks and blendshape coefficients in the browser at video rate, which is enough to drive an avatar's face from a webcam.",+ "tags": [+ "oss",+ "js",+ "python",+ "tracking"+ ]+ },+ {+ "name": "ARKit blendshapes",+ "url": "https://developer.apple.com/documentation/arkit/arfaceanchor/blendshapelocation",+ "description": "The 52-coefficient facial vocabulary that avatar formats, capture tools, and generators have all standardised on.",+ "tags": [+ "spec",+ "reference"+ ]+ },+ {+ "name": "Whisper",+ "url": "https://github.com/openai/whisper",+ "description": "Robust multilingual speech recognition that made self-hosted transcription the default rather than the exception.",+ "tags": [+ "oss",+ "python",+ "stt",+ "foundational"+ ]+ },+ {+ "name": "faster-whisper",+ "url": "https://github.com/SYSTRAN/faster-whisper",+ "description": "A CTranslate2 reimplementation of Whisper that is several times faster at the same accuracy. Use this in production, not the reference code.",+ "tags": [+ "oss",+ "python",+ "stt"+ ]+ },+ {+ "name": "whisper.cpp",+ "url": "https://github.com/ggml-org/whisper.cpp",+ "description": "Whisper in plain C and C++ with no dependencies, running on phones, browsers via WebAssembly, and Raspberry Pis.",+ "tags": [+ "oss",+ "cpp",+ "stt",+ "edge"+ ]+ },+ {+ "name": "Kokoro",+ "url": "https://github.com/hexgrad/kokoro",+ "description": "An 82M-parameter text to speech model with quality that embarrasses far larger ones, small enough to run in a browser tab.",+ "tags": [+ "oss",+ "tts",+ "small"+ ]+ },+ {+ "name": "Piper",+ "url": "https://github.com/rhasspy/piper",+ "description": "Fast local neural text to speech built for low-power devices, with a large set of pretrained voices.",+ "tags": [+ "oss",+ "tts",+ "edge"+ ]+ },+ {+ "name": "F5-TTS",+ "url": "https://github.com/SWivid/F5-TTS",+ "description": "Flow-matching text to speech with convincing zero-shot voice cloning from a few seconds of reference audio.",+ "tags": [+ "oss",+ "tts",+ "cloning"+ ]+ },+ {+ "name": "Coqui TTS",+ "url": "https://github.com/coqui-ai/TTS",+ "description": "A deep-learning toolkit covering dozens of text to speech and voice-conversion models, including XTTS multilingual cloning.",+ "tags": [+ "oss",+ "python",+ "tts"+ ]+ },+ {+ "name": "Silero VAD",+ "url": "https://github.com/snakers4/silero-vad",+ "description": "Voice activity detection in under a millisecond per chunk. The unglamorous component that decides whether a voice agent feels responsive.",+ "tags": [+ "oss",+ "vad",+ "small"+ ]+ },+ {+ "name": "Pipecat",+ "url": "https://github.com/pipecat-ai/pipecat",+ "description": "A Python framework for real-time voice and multimodal agents: pipelines, interruption handling, and transport out of the box.",+ "tags": [+ "oss",+ "python",+ "realtime"+ ]+ },+ {+ "name": "LiveKit Agents",+ "url": "https://github.com/livekit/agents",+ "description": "Build voice agents that join a WebRTC room, with turn detection and telephony handled for you.",+ "tags": [+ "oss",+ "python",+ "realtime",+ "webrtc"+ ]+ },+ {+ "name": "Web Speech API",+ "url": "https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API",+ "description": "Recognition and synthesis already in the browser, with zero bytes shipped. Always check this before adding a model.",+ "tags": [+ "spec",+ "browser",+ "free"+ ]+ }+ ]+ },+ {+ "id": "renderers",+ "title": "Rendering on the web",+ "description": "The runtime that actually puts the body on a page. Everything above is upstream of a choice made here.",+ "items": [+ {+ "name": "three.js",+ "url": "https://github.com/mrdoob/three.js",+ "description": "The default WebGL and WebGPU library for the web, with the deepest ecosystem of loaders, controls, and examples anywhere.",+ "tags": [+ "oss",+ "js",+ "foundational"+ ]+ },+ {+ "name": "React Three Fiber",+ "url": "https://github.com/pmndrs/react-three-fiber",+ "description": "three.js as a React renderer. Scene graphs become components, and the reconciler handles the mutation you would otherwise write by hand.",+ "tags": [+ "oss",+ "js",+ "react",+ "threejs"+ ]+ },+ {+ "name": "drei",+ "url": "https://github.com/pmndrs/drei",+ "description": "The helper library that makes React Three Fiber practical: controls, loaders, staging, text, and effects you would otherwise rebuild every project.",+ "tags": [+ "oss",+ "js",+ "react"+ ]+ },+ {+ "name": "Babylon.js",+ "url": "https://github.com/BabylonJS/Babylon.js",+ "description": "A batteries-included web engine with a built-in physics layer, node material editor, and an inspector that is genuinely better than the alternatives.",+ "tags": [+ "oss",+ "js",+ "engine"+ ]+ },+ {+ "name": "PlayCanvas",+ "url": "https://github.com/playcanvas/engine",+ "description": "A lightweight engine tuned hard for load time and mobile, backed by a collaborative browser editor.",+ "tags": [+ "oss",+ "js",+ "engine"+ ]+ },+ {+ "name": "model-viewer",+ "url": "https://github.com/google/model-viewer",+ "description": "One custom element that renders a glTF with sensible defaults, accessibility, and AR on both phone platforms. The lowest-effort correct answer.",+ "tags": [+ "oss",+ "js",+ "webcomponent",+ "ar"+ ]+ },+ {+ "name": "A-Frame",+ "url": "https://github.com/aframevr/aframe",+ "description": "Declarative HTML for WebXR scenes, with an entity-component system underneath. Still the shortest path from markup to headset.",+ "tags": [+ "oss",+ "js",+ "webxr"+ ]+ },+ {+ "name": "Threlte",+ "url": "https://github.com/threlte/threlte",+ "description": "three.js for Svelte, with a component library, physics bindings, and a preprocessor that keeps bundles small.",+ "tags": [+ "oss",+ "js",+ "svelte"+ ]+ },+ {+ "name": "TresJS",+ "url": "https://github.com/Tresjs/tres",+ "description": "three.js as Vue components, with reactivity mapped onto scene-graph updates.",+ "tags": [+ "oss",+ "js",+ "vue"+ ]+ },+ {+ "name": "OGL",+ "url": "https://github.com/oframe/ogl",+ "description": "A minimal WebGL library that stays close to the API instead of hiding it. The right tool when three.js is more engine than you need.",+ "tags": [+ "oss",+ "js",+ "minimal"+ ]+ },+ {+ "name": "Needle Engine",+ "url": "https://needle.tools",+ "description": "Author in Unity or Blender, ship a small web build. A pragmatic bridge for teams whose content pipeline already exists.",+ "tags": [+ "service",+ "js",+ "pipeline"+ ]+ },+ {+ "name": "agent-3d",+ "url": "https://github.com/nirholas/three.ws/tree/main/avatar-sdk",+ "description": "A web component that drops a rigged, animated, lip-syncing agent avatar onto any page in one tag, with a JS API for moods and gestures.",+ "tags": [+ "oss",+ "js",+ "webcomponent",+ "avatars"+ ]+ }+ ]+ },+ {+ "id": "formats",+ "title": "Formats and asset pipeline",+ "description": "Getting the asset small, valid, and loadable. Most 3D-on-the-web failures are pipeline failures, not rendering failures.",+ "items": [+ {+ "name": "glTF",+ "url": "https://github.com/KhronosGroup/glTF",+ "description": "The transmission format for 3D on the web, plus its extension registry. If you are choosing a format, this is the answer.",+ "tags": [+ "spec",+ "foundational"+ ]+ },+ {+ "name": "glTF-Transform",+ "url": "https://github.com/donmccurdy/glTF-Transform",+ "description": "Read, edit, optimise, and validate glTF from Node or the CLI. Deduplicate, resize textures, weld, and compress in a scripted pipeline.",+ "tags": [+ "oss",+ "js",+ "cli",+ "pipeline"+ ]+ },+ {+ "name": "meshoptimizer",+ "url": "https://github.com/zeux/meshoptimizer",+ "description": "Vertex cache optimisation, simplification, and the meshopt compression that glTF files use. Its gltfpack CLI is the single highest-leverage step in most pipelines.",+ "tags": [+ "oss",+ "cpp",+ "compression",+ "foundational"+ ]+ },+ {+ "name": "Draco",+ "url": "https://github.com/google/draco",+ "description": "Geometry compression that shrinks meshes and point clouds hard, with decoders for the browser.",+ "tags": [+ "oss",+ "cpp",+ "compression"+ ]+ },+ {+ "name": "KTX-Software",+ "url": "https://github.com/KhronosGroup/KTX-Software",+ "description": "Basis Universal texture transcoding. GPU-compressed textures that stay compressed in memory, which is usually the real budget problem.",+ "tags": [+ "oss",+ "textures",+ "compression"+ ]+ },+ {+ "name": "glTF-Validator",+ "url": "https://github.com/KhronosGroup/glTF-Validator",+ "description": "Tells you whether a file is actually valid before a user's browser does. Wire it into CI and stop guessing.",+ "tags": [+ "oss",+ "validation",+ "ci"+ ]+ },+ {+ "name": "glTF Sample Assets",+ "url": "https://github.com/KhronosGroup/glTF-Sample-Assets",+ "description": "The official test models, including deliberately awkward ones. Your loader is not done until these render correctly.",+ "tags": [+ "assets",+ "testing"+ ]+ },+ {+ "name": "Blender",+ "url": "https://www.blender.org",+ "description": "Free, scriptable in Python, and headless-capable, which makes it the workhorse of automated 3D pipelines as much as an authoring tool.",+ "tags": [+ "oss",+ "free",+ "desktop",+ "python"+ ]+ },+ {+ "name": "trimesh",+ "url": "https://github.com/mikedh/trimesh",+ "description": "The Python library for loading, repairing, measuring, and booleaning meshes. Nearly every 3D backend service has this in it somewhere.",+ "tags": [+ "oss",+ "python",+ "foundational"+ ]+ },+ {+ "name": "Assimp",+ "url": "https://github.com/assimp/assimp",+ "description": "Imports roughly fifty 3D formats into one data structure. The escape hatch when someone sends you a file nothing else opens.",+ "tags": [+ "oss",+ "cpp",+ "conversion"+ ]+ },+ {+ "name": "OpenUSD",+ "url": "https://github.com/PixarAnimationStudios/OpenUSD",+ "description": "Pixar's scene description for large, layered, collaboratively edited worlds. Heavier than glTF, and the right answer above a certain scale.",+ "tags": [+ "oss",+ "cpp",+ "spec"+ ]+ },+ {+ "name": "three-mesh-bvh",+ "url": "https://github.com/gkjohnson/three-mesh-bvh",+ "description": "Bounding volume hierarchies for three.js geometry, turning raycasts against dense meshes from unusable into instant.",+ "tags": [+ "oss",+ "js",+ "threejs",+ "performance"+ ]+ }+ ]+ },+ {+ "id": "worlds",+ "title": "Physics, networking, and shared worlds",+ "description": "One avatar in a viewer is a demo. Several avatars in a place, with gravity and collisions, is a product.",+ "items": [+ {+ "name": "Rapier",+ "url": "https://github.com/dimforge/rapier",+ "description": "A Rust physics engine with deterministic cross-platform simulation and first-class WebAssembly bindings.",+ "tags": [+ "oss",+ "rust",+ "wasm",+ "physics"+ ]+ },+ {+ "name": "Jolt Physics",+ "url": "https://github.com/jrouwe/JoltPhysics",+ "description": "The multi-core physics engine used in shipped AAA titles, with a WebAssembly build that runs in the browser.",+ "tags": [+ "oss",+ "cpp",+ "wasm",+ "physics"+ ]+ },+ {+ "name": "cannon-es",+ "url": "https://github.com/pmndrs/cannon-es",+ "description": "A maintained fork of cannon.js in modern JavaScript. Small and readable, and enough for character controllers and props.",+ "tags": [+ "oss",+ "js",+ "physics"+ ]+ },+ {+ "name": "Colyseus",+ "url": "https://github.com/colyseus/colyseus",+ "description": "Authoritative multiplayer rooms with automatic state synchronisation and delta encoding, in Node.",+ "tags": [+ "oss",+ "js",+ "multiplayer"+ ]+ },+ {+ "name": "Yjs",+ "url": "https://github.com/yjs/yjs",+ "description": "CRDTs for shared editable state, with providers for WebRTC, WebSocket, and IndexedDB persistence.",+ "tags": [+ "oss",+ "js",+ "crdt",+ "collaboration"+ ]+ },+ {+ "name": "LiveKit",+ "url": "https://github.com/livekit/livekit",+ "description": "An open WebRTC server for real-time audio and video, with SDKs everywhere and a self-hostable path.",+ "tags": [+ "oss",+ "go",+ "webrtc"+ ]+ },+ {+ "name": "mediasoup",+ "url": "https://github.com/versatica/mediasoup",+ "description": "A selective forwarding unit as a Node library rather than a service, for teams that want to own the media path.",+ "tags": [+ "oss",+ "js",+ "webrtc"+ ]+ },+ {+ "name": "geckos.io",+ "url": "https://github.com/geckosio/geckos.io",+ "description": "Unreliable UDP-like messaging in the browser over WebRTC data channels, which is what fast-moving avatar positions actually want.",+ "tags": [+ "oss",+ "js",+ "networking"+ ]+ },+ {+ "name": "WebXR Device API",+ "url": "https://immersive-web.github.io/webxr/",+ "description": "The specification behind every headset and phone AR session on the open web. Read the spec, not just a framework's wrapper.",+ "tags": [+ "spec",+ "webxr"+ ]+ }+ ]+ },+ {+ "id": "brains",+ "title": "Agent frameworks and protocols",+ "description": "The mind that drives the body. Listed here because embodiment is a rendering problem bolted onto an agent problem, and most teams underestimate the second one.",+ "items": [+ {+ "name": "Model Context Protocol",+ "url": "https://github.com/modelcontextprotocol",+ "description": "The open protocol for giving a model tools, resources, and prompts from an external server. The cleanest way to let an agent act on a 3D pipeline.",+ "tags": [+ "spec",+ "oss",+ "tools"+ ]+ },+ {+ "name": "Claude Agent SDK",+ "url": "https://github.com/anthropics/claude-agent-sdk-python",+ "description": "Build agents on the same harness Claude Code runs on: tool loops, subagents, permissions, and session state.",+ "tags": [+ "oss",+ "python",+ "sdk"+ ]+ },+ {+ "name": "OpenAI Agents SDK",+ "url": "https://github.com/openai/openai-agents-python",+ "description": "A small framework for multi-agent handoffs, guardrails, and tracing, with a deliberately thin abstraction layer.",+ "tags": [+ "oss",+ "python",+ "sdk"+ ]+ },+ {+ "name": "LangGraph",+ "url": "https://github.com/langchain-ai/langgraph",+ "description": "Agents as explicit stateful graphs, with checkpointing and human-in-the-loop pauses. The right shape when a flow must be resumable.",+ "tags": [+ "oss",+ "python",+ "js",+ "framework"+ ]+ },+ {+ "name": "Vercel AI SDK",+ "url": "https://github.com/vercel/ai",+ "description": "One TypeScript interface across model providers, with streaming, structured output, and tool calls that work in the browser.",+ "tags": [+ "oss",+ "js",+ "streaming"+ ]+ },+ {+ "name": "smolagents",+ "url": "https://github.com/huggingface/smolagents",+ "description": "Agents that write Python to call their tools instead of emitting JSON, in about a thousand lines of library code.",+ "tags": [+ "oss",+ "python",+ "minimal"+ ]+ },+ {+ "name": "CrewAI",+ "url": "https://github.com/crewAIInc/crewAI",+ "description": "Role-based multi-agent orchestration with an emphasis on readable task delegation.",+ "tags": [+ "oss",+ "python",+ "multi-agent"+ ]+ },+ {+ "name": "AutoGen",+ "url": "https://github.com/microsoft/autogen",+ "description": "A research-grade framework for conversational multi-agent systems, with an event-driven core and a visual builder.",+ "tags": [+ "oss",+ "python",+ "multi-agent"+ ]+ },+ {+ "name": "Mastra",+ "url": "https://github.com/mastra-ai/mastra",+ "description": "A TypeScript agent framework with workflows, memory, and evals, aimed at shipping rather than research.",+ "tags": [+ "oss",+ "js",+ "framework"+ ]+ },+ {+ "name": "Letta",+ "url": "https://github.com/letta-ai/letta",+ "description": "Agents with persistent memory as a first-class server concept, out of the MemGPT research line.",+ "tags": [+ "oss",+ "python",+ "memory"+ ]+ },+ {+ "name": "A2A",+ "url": "https://github.com/a2aproject/A2A",+ "description": "An open protocol for agents built by different vendors to discover each other and collaborate over a common envelope.",+ "tags": [+ "spec",+ "interop"+ ]+ },+ {+ "name": "Generative Agents",+ "url": "https://github.com/joonspk-research/generative_agents",+ "description": "The Stanford simulation where twenty-five agents with memory, reflection, and planning produced emergent social behaviour. Still the clearest demonstration of why memory architecture matters.",+ "tags": [+ "research",+ "foundational",+ "simulation"+ ]+ },+ {+ "name": "Voyager",+ "url": "https://github.com/MineDojo/Voyager",+ "description": "An embodied agent that plays Minecraft by writing and saving its own skills as code, then composing them. The best existing argument for embodiment as a learning signal.",+ "tags": [+ "research",+ "embodied",+ "foundational"+ ]+ }+ ]+ },+ {+ "id": "datasets",+ "title": "Datasets and simulators",+ "description": "What the models in this list were trained and evaluated on, and where embodied agents go to practise.",+ "items": [+ {+ "name": "Objaverse",+ "url": "https://objaverse.allenai.org",+ "description": "Around 800,000 annotated 3D objects, and the XL release pushes past ten million. The dataset that made open 3D generation viable.",+ "tags": [+ "dataset",+ "free",+ "foundational"+ ]+ },+ {+ "name": "Objaverse-XL",+ "url": "https://github.com/allenai/objaverse-xl",+ "description": "The download and processing tooling for the ten-million-object release, including deduplication and rendering scripts.",+ "tags": [+ "dataset",+ "python"+ ]+ },+ {+ "name": "ShapeNet",+ "url": "https://huggingface.co/datasets/ShapeNet/ShapeNetCore",+ "description": "The categorised CAD-model dataset that anchored 3D deep learning for a decade, mirrored on Hugging Face while its original site is unreliable.",+ "tags": [+ "dataset",+ "research"+ ]+ },+ {+ "name": "Habitat",+ "url": "https://github.com/facebookresearch/habitat-lab",+ "description": "A fast photorealistic simulator for training embodied agents to navigate and manipulate in indoor scenes.",+ "tags": [+ "oss",+ "python",+ "simulator"+ ]+ },+ {+ "name": "AI2-THOR",+ "url": "https://github.com/allenai/ai2thor",+ "description": "Interactive 3D rooms where objects have state: doors open, eggs break, appliances turn on. Built for agents that change the world rather than tour it.",+ "tags": [+ "oss",+ "python",+ "simulator"+ ]+ },+ {+ "name": "BEHAVIOR",+ "url": "https://behavior.stanford.edu",+ "description": "A benchmark of a thousand everyday household activities defined in a logic language, with simulated physics and states.",+ "tags": [+ "benchmark",+ "research"+ ]+ },+ {+ "name": "ScanNet",+ "url": "http://www.scan-net.org",+ "description": "Richly annotated RGB-D scans of real indoor spaces, and the benchmark most scene-understanding work still reports on.",+ "tags": [+ "dataset",+ "research"+ ]+ }+ ]+ },+ {+ "id": "assets",+ "title": "Free assets",+ "description": "Models, textures, and lighting you can ship without a lawyer. Every project in this list looks better with a real HDRI behind it.",+ "items": [+ {+ "name": "Poly Haven",+ "url": "https://polyhaven.com",+ "description": "HDRIs, PBR textures, and models, all CC0, all high quality, with no account required. Start here.",+ "tags": [+ "free",+ "cc0",+ "hdri",+ "textures"+ ]+ },+ {+ "name": "ambientCG",+ "url": "https://ambientcg.com",+ "description": "Thousands of CC0 PBR material sets at multiple resolutions, downloadable in bulk.",+ "tags": [+ "free",+ "cc0",+ "textures"+ ]+ },+ {+ "name": "Kenney",+ "url": "https://kenney.nl",+ "description": "Tens of thousands of CC0 game assets, including clean low-poly 3D kits that prototype beautifully.",+ "tags": [+ "free",+ "cc0",+ "models"+ ]+ },+ {+ "name": "Quaternius",+ "url": "https://quaternius.com",+ "description": "Stylised CC0 model packs with consistent proportions across sets, so a scene assembled from several packs still looks intentional.",+ "tags": [+ "free",+ "cc0",+ "models"+ ]+ },+ {+ "name": "Sketchfab",+ "url": "https://sketchfab.com",+ "description": "The largest 3D model library on the web, with a downloadable Creative Commons subset and glTF export.",+ "tags": [+ "service",+ "models"+ ]+ },+ {+ "name": "OpenGameArt",+ "url": "https://opengameart.org",+ "description": "A long-running community archive of openly licensed art. Uneven, and it holds things that exist nowhere else.",+ "tags": [+ "free",+ "models",+ "community"+ ]+ }+ ]+ },+ {+ "id": "learn",+ "title": "Learning",+ "description": "How to get good at this. Read in roughly this order.",+ "items": [+ {+ "name": "three.js manual",+ "url": "https://threejs.org/manual/",+ "description": "The official guide, rewritten around real problems rather than API surface. The fastest correct start.",+ "tags": [+ "free",+ "docs"+ ]+ },+ {+ "name": "Discover three.js",+ "url": "https://discoverthreejs.com",+ "description": "A free book that teaches the concepts underneath the library, including colour management and performance work most tutorials skip.",+ "tags": [+ "free",+ "book"+ ]+ },+ {+ "name": "Three.js Journey",+ "url": "https://threejs-journey.com",+ "description": "Bruno Simon's paid course, and the most thorough path from nothing to shaders and production scenes.",+ "tags": [+ "course",+ "paid"+ ]+ },+ {+ "name": "The Book of Shaders",+ "url": "https://thebookofshaders.com",+ "description": "An interactive introduction to fragment shaders. Work through it once and materials stop being magic.",+ "tags": [+ "free",+ "shaders",+ "interactive"+ ]+ },+ {+ "name": "WebGL Fundamentals",+ "url": "https://webglfundamentals.org",+ "description": "What every abstraction in this list is hiding, explained from first principles with runnable examples.",+ "tags": [+ "free",+ "webgl"+ ]+ },+ {+ "name": "WebGPU Fundamentals",+ "url": "https://webgpufundamentals.org",+ "description": "The same treatment for WebGPU, which is where the web's rendering stack is heading.",+ "tags": [+ "free",+ "webgpu"+ ]+ },+ {+ "name": "Real-Time Rendering resources",+ "url": "https://www.realtimerendering.com",+ "description": "The companion site to the standard reference text, with a maintained portal into graphics literature.",+ "tags": [+ "free",+ "reference",+ "graphics"+ ]+ }+ ]+ }+ ]+}diff --git a/public/changelog.json b/public/changelog.jsonindex 994e6e3ae..44d907342 100644--- a/public/changelog.json+++ b/public/changelog.json@@ -1,11 +1,21 @@{- "generated_at": "2026-09-07T20:49:20.490Z",+ "generated_at": "2026-09-07T21:42:43.066Z","generated_by": "scripts/build-page-index.mjs from data/pages.json + data/changelog.json","site": {"name": "three.ws","url": "https://three.ws"},"entries": [+ {+ "date": "2026-09-07",+ "type": "launch",+ "title": "Awesome 3D Agents",+ "summary": "A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.",+ "link": "/awesome",+ "tags": [+ "launch"+ ]+ },{"date": "2026-09-07","type": "launch",@@ -70,6 +80,17 @@"infra"]},+ {+ "date": "2026-09-07",+ "type": "update",+ "title": "Awesome 3D Agents: a curated list of everything you need to give an AI a body",+ "summary": "Giving an agent a body means assembling a pipeline out of a dozen separate fields, and the map of that pipeline did not exist anywhere. There is now one: 134 hand-picked projects across 13 sections, covering text-to-3D generation, photo and video capture, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, physics and networking, agent frameworks, datasets, free assets, and where to learn all of it. Each entry says in one sentence what the thing does and why you would reach for it instead of the alternative. The list is searchable and filterable by tag at /awesome, copyable as Markdown, and published in the standard awesome-list format on GitHub so it can be read, forked, and contributed to there. Every link is fetched and verified before it ships.",+ "link": "/awesome",+ "tags": [+ "feature",+ "docs"+ ]+ },{"date": "2026-09-07","type": "update",diff --git a/public/changelog.xml b/public/changelog.xmlindex 5869da7ed..dfc082dd7 100644--- a/public/changelog.xml+++ b/public/changelog.xml@@ -6,6 +6,14 @@<atom:link href="https://three.ws/changelog.xml" rel="self" type="application/rss+xml"/><description>What's new on three.ws: features, improvements, and releases, newest first.</description><language>en</language>+ <item>+ <title>Awesome 3D Agents</title>+ <link>https://three.ws/awesome</link>+ <guid isPermaLink="false">2026-09-07:Awesome 3D Agents</guid>+ <pubDate>Mon, 07 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.</description>+ </item><item><title>Docs · SEO and crawler rendering</title><link>https://three.ws/docs/seo</link>@@ -54,6 +62,14 @@<category>fix</category><description>Google Search Console flagged an archived story page for unparsable structured data. The cause: the archive trims a publisher summary to a fixed length before serving it, and the cut landed in the middle of an emoji, leaving half a character in the page data that search engines read. Trimming is now character-aware, so it never splits an emoji, a flag, or an accented letter, and the page data is checked for half characters before it is written. Around one archived story in seven thousand carried the flaw; all of them serve a valid listing now. The same fix covers the story summaries shown in link previews and the shared page descriptions across the site.</description></item>+ <item>+ <title>Awesome 3D Agents: a curated list of everything you need to give an AI a body</title>+ <link>https://three.ws/awesome</link>+ <guid isPermaLink="false">2026-09-07:Awesome 3D Agents: a curated list of everything you need to give an AI a body</guid>+ <pubDate>Mon, 07 Sep 2026 12:00:00 GMT</pubDate>+ <category>feature</category>+ <description>Giving an agent a body means assembling a pipeline out of a dozen separate fields, and the map of that pipeline did not exist anywhere. There is now one: 134 hand-picked projects across 13 sections, covering text-to-3D generation, photo and video capture, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, physics and networking, agent frameworks, datasets, free assets, and where to learn all of it. Each entry says in one sentence what the thing does and why you would reach for it instead of the alternative. The list is searchable and filterable by tag at /awesome, copyable as Markdown, and published in the standard awesome-list format on GitHub so it can be read, forked, and contributed to there. Every link is fetched and verified before it ships.</description>+ </item><item><title>Blender tools now open three.ws avatars, which Blender alone cannot</title><link>https://three.ws/changelog</link>diff --git a/public/features.json b/public/features.jsonindex 990d4a5b4..0b806878e 100644--- a/public/features.json+++ b/public/features.json@@ -1,5 +1,5 @@{- "generated_at": "2026-09-07T20:35:47.664Z",+ "generated_at": "2026-09-07T21:42:43.055Z","generated_by": "scripts/build-page-index.mjs from data/pages.json","site": {"name": "three.ws",@@ -7497,6 +7497,16 @@"indexable": true,"tags": [],"showcase": false+ },+ {+ "path": "/awesome",+ "title": "Awesome 3D Agents",+ "description": "A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.",+ "added": "2026-09-07",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": true}]},diff --git a/public/llms-full.txt b/public/llms-full.txtindex 5f14c80fe..9cc343406 100644--- a/public/llms-full.txt+++ b/public/llms-full.txt@@ -4546,6 +4546,12 @@ URL: https://three.ws/docs/pocketHow the /pocket handheld works: the D-pad and gamepad controls, the three cartridges, the URL params that make a share link, and the one iframe that puts a live 3D agent console on any site.+### Awesome 3D Agents++URL: https://three.ws/awesome++A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.+## BlogLong-form announcements, launches, and writing from three.ws.diff --git a/public/llms.txt b/public/llms.txtindex 54ea55c01..5095a6492 100644--- a/public/llms.txt+++ b/public/llms.txt@@ -773,6 +773,7 @@ Docs, tutorials, and reference.- [Docs · Terminal renderer](https://three.ws/docs/tty): curl three.ws/tty and a rigged 3D agent walks in your terminal. The hosted streaming endpoint and its parameters, plus @three-ws/tty-3d, the CLI and library that render any GLB in a terminal with a pure-CPU rasterizer: real skeletal animation, half-block pixels, no GPU and no browser.- [Tutorial · Connect your home](https://three.ws/tutorials/connect-your-home): Point your agent at a real Home Assistant and it runs your actual house: lights, thermostat, and the scenes you already built. Covers the three honest ways to reach a house that lives on a home network, and the rule that unlocking, opening and disarming always stop and ask, while locking up never...- [Docs · Pocket Console](https://three.ws/docs/pocket): How the /pocket handheld works: the D-pad and gamepad controls, the three cartridges, the URL params that make a share link, and the one iframe that puts a live 3D agent console on any site.+- [Awesome 3D Agents](https://three.ws/awesome): A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.## Blogdiff --git a/public/nav-data.js b/public/nav-data.jsindex 4db6216b0..48097528c 100644--- a/public/nav-data.js+++ b/public/nav-data.js@@ -736,6 +736,11 @@ export const NAV_GROUPS = [{ title: 'Tutorials', href: '/tutorials', desc: 'Step-by-step guides' },{ title: 'Examples', href: '/examples', desc: 'Runnable copy-paste code' },{ title: 'Cookbook', href: '/cookbook', desc: 'Recipes you download and run' },+ {+ title: 'Awesome 3D Agents',+ href: '/awesome',+ desc: 'The curated list of tools for giving an AI a body',+ },{ title: 'Chat', href: '/chat', desc: 'Talk to your agent' },],},diff --git a/public/sitemap/index.html b/public/sitemap/index.htmlindex 5a328f145..956ebcc4a 100644--- a/public/sitemap/index.html+++ b/public/sitemap/index.html@@ -88,7 +88,7 @@<main class="sm-wrap"><div class="sm-hero"><h1>Sitemap</h1>- <p>All 916 pages on three.ws, grouped by purpose. Looking for the machine-readable versions?</p>+ <p>All 917 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>@@ -100,7 +100,7 @@</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 916 pages by name, path, or description…" aria-label="Filter pages" autocomplete="off" spellcheck="false" />+ <input id="sm-filter-input" type="search" placeholder="Filter 917 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>@@ -116,6 +116,14 @@<span class="sm-desc">How a JavaScript site is served to search engines: the four server-side render paths, the contract that keeps a page and its sitemap entry agreeing, and how to audit the live origin.</span></a></li>+ <li>+ <a href="/awesome">+ <span class="sm-new-date">Sep 7, 2026</span>+ <span class="sm-title">Awesome 3D Agents</span>+ <span class="sm-path">/awesome</span>+ <span class="sm-desc">A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.</span>+ </a>+ </li><li><a href="/irl/sign"><span class="sm-new-date">Sep 5, 2026</span>@@ -172,14 +180,6 @@<span class="sm-desc">The season recap for any verified pump.fun trading agent on three.ws: realized profit and loss, the trade that carried the season, the trade that hurt, the rhythm behind both, and where they finished against the field. Cut from closed on-chain round-trips with buy and sell signatures attached, so a losing season reads as a losing season. Swipeable, shareable, no account and no wallet.</span></a></li>- <li>- <a href="/drive">- <span class="sm-new-date">Sep 3, 2026</span>- <span class="sm-title">three.ws Drive</span>- <span class="sm-path">/drive</span>- <span class="sm-desc">Your 3D agent in the car: voice-first, glanceable, and built to the rules CarPlay and Android Auto enforce. Runs on a head unit, a phone in the cradle, or inside the native car shell.</span>- </a>- </li></ul></section><nav class="sm-toc" aria-label="Sections">@@ -2505,7 +2505,7 @@<section class="sm-section" id="learn"><header>- <h2>Learn<span class="sm-count" data-count>422</span></h2>+ <h2>Learn<span class="sm-count" data-count>423</span></h2><p class="sm-section-desc">Docs, tutorials, and reference.</p></header><ul class="sm-list">@@ -5463,6 +5463,13 @@<span class="sm-desc">How the /pocket handheld works: the D-pad and gamepad controls, the three cartridges, the URL params that make a share link, and the one iframe that puts a live 3D agent console on any site.</span></a></li>+ <li>+ <a href="/awesome">+ <span class="sm-title">Awesome 3D Agents</span>+ <span class="sm-path">/awesome</span>+ <span class="sm-desc">A curated, link-checked list of everything you need to give an AI agent a body: text-to-3D models, capture and Gaussian splatting, avatars, rigging, motion generation, lipsync and voice, web renderers, asset pipelines, and the agent frameworks that drive them. Searchable and filterable, and published as an awesome list on GitHub.</span>+ </a>+ </li></ul></section>diff --git a/scripts/build-awesome.mjs b/scripts/build-awesome.mjsnew file mode 100644index 000000000..508db7e03--- /dev/null+++ b/scripts/build-awesome.mjs@@ -0,0 +1,163 @@+#!/usr/bin/env node+/**+ * Generate the Awesome 3D Agents list from data/awesome.json.+ *+ * One source, two outputs, so the GitHub-facing list and the browsable page on+ * three.ws can never disagree:+ * awesome/README.md the list, in the awesome-list format people expect+ * public/awesome.json the same data plus counts and a tag index, for /awesome+ *+ * node scripts/build-awesome.mjs write both+ * node scripts/build-awesome.mjs --check fail if either is out of date+ *+ * The --check mode runs in `npm run gate`, so an edit to the data that never+ * got built cannot ship a stale README.+ */+import { mkdirSync, readFileSync, writeFileSync } from 'node:fs';+import path from 'node:path';+import { fileURLToPath } from 'node:url';++const root = path.resolve(path.dirname(fileURLToPath(import.meta.url)), '..');+const check = process.argv.includes('--check');++const list = JSON.parse(readFileSync(path.join(root, 'data/awesome.json'), 'utf8'));++const problems = [];+const slugs = new Set();+const urls = new Map();+for (const section of list.sections) {+ if (slugs.has(section.id)) problems.push(`duplicate section id: ${section.id}`);+ slugs.add(section.id);+ if (!section.items.length) problems.push(`empty section: ${section.id}`);+ for (const item of section.items) {+ for (const field of ['name', 'url', 'description']) {+ if (!item[field]) problems.push(`${section.id}/${item.name ?? '?'} is missing ${field}`);+ }+ if (item.url && !/^https?:\/\//.test(item.url)) problems.push(`${item.name}: url must be absolute`);+ // The awesome format renders one sentence per entry. A description that+ // runs past this reads as a paragraph in a list and breaks the scan.+ if (item.description && item.description.length > 260) {+ problems.push(`${item.name}: description is ${item.description.length} chars, keep it under 260`);+ }+ if (/[\u2014\u2013]/.test(`${item.name}${item.description}`)) {+ problems.push(`${item.name}: uses a dash character that is banned in this repo`);+ }+ const key = item.url?.replace(/\/+$/, '').toLowerCase();+ if (key && urls.has(key)) problems.push(`${item.name} repeats the url of ${urls.get(key)}`);+ else if (key) urls.set(key, item.name);+ }+}+if (problems.length) {+ for (const p of problems) console.error(`[build-awesome] ${p}`);+ process.exit(1);+}++const total = list.sections.reduce((n, s) => n + s.items.length, 0);++// ── awesome/README.md ────────────────────────────────────────────────────────+const anchor = (title) =>+ title+ .toLowerCase()+ .replace(/[^\w\s-]/g, '')+ .trim()+ .replace(/\s+/g, '-');++const md = [];+md.push(`# ${list.title}`);+md.push('');+md.push('[](https://awesome.re)');+md.push('');+md.push(`> ${list.tagline}`);+md.push('');+md.push(list.intro);+md.push('');+md.push(+ `${total} entries across ${list.sections.length} sections. Browsable, searchable, and filterable at [${list.page.replace(/^https?:\/\//, '')}](${list.page}). Every link is checked with \`npm run awesome:links\`.`,+);+md.push('');+md.push('## Contents');+md.push('');+for (const section of list.sections) {+ md.push(`- [${section.title}](#${anchor(section.title)}) (${section.items.length})`);+}+md.push('');+for (const section of list.sections) {+ md.push(`## ${section.title}`);+ md.push('');+ md.push(section.description);+ md.push('');+ for (const item of section.items) {+ md.push(`- [${item.name}](${item.url}) - ${item.description}`);+ }+ md.push('');+}+md.push('## Contributing');+md.push('');+md.push(+ 'Additions are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md) first: entries live in [`data/awesome.json`](../data/awesome.json), not in this file, which is generated.',+);+md.push('');+md.push('## License');+md.push('');+md.push(+ 'The list is published under [Apache-2.0](../LICENSE) with the rest of this repository. Each linked project carries its own license.',+);+md.push('');+const readme = md.join('\n');++// ── public/awesome.json ──────────────────────────────────────────────────────+const tagCounts = new Map();+for (const section of list.sections) {+ for (const item of section.items) {+ for (const tag of item.tags ?? []) tagCounts.set(tag, (tagCounts.get(tag) ?? 0) + 1);+ }+}+const payload = {+ title: list.title,+ tagline: list.tagline,+ intro: list.intro,+ repo: `${list.repo}/tree/main/awesome`,+ markdown: `${list.repo}/blob/main/awesome/README.md`,+ counts: { items: total, sections: list.sections.length, tags: tagCounts.size },+ // Tags the page offers as filter chips, most common first. A tag used once+ // is noise in a filter bar, so it stays searchable but is not offered.+ tags: [...tagCounts.entries()]+ .filter(([, n]) => n > 1)+ .sort((a, b) => b[1] - a[1] || a[0].localeCompare(b[0]))+ .map(([id, count]) => ({ id, count })),+ sections: list.sections.map((s) => ({+ id: s.id,+ title: s.title,+ description: s.description,+ items: s.items.map((i) => ({ ...i, tags: i.tags ?? [] })),+ })),+};+const json = `${JSON.stringify(payload, null, '\t')}\n`;++const outputs = [+ { file: path.join(root, 'awesome/README.md'), content: readme },+ { file: path.join(root, 'public/awesome.json'), content: json },+];++if (check) {+ const stale = outputs.filter((o) => {+ try {+ return readFileSync(o.file, 'utf8') !== o.content;+ } catch {+ return true;+ }+ });+ if (stale.length) {+ for (const o of stale) console.error(`[build-awesome] out of date: ${path.relative(root, o.file)}`);+ console.error('[build-awesome] run `npm run build:awesome`');+ process.exit(1);+ }+ console.log(`[build-awesome] up to date (${total} entries)`);+} else {+ for (const o of outputs) {+ mkdirSync(path.dirname(o.file), { recursive: true });+ writeFileSync(o.file, o.content);+ console.log(`[build-awesome] wrote ${path.relative(root, o.file)}`);+ }+ console.log(`[build-awesome] ${total} entries across ${list.sections.length} sections`);+}diff --git a/scripts/check-awesome-links.mjs b/scripts/check-awesome-links.mjsnew file mode 100644index 000000000..cb2627761--- /dev/null+++ b/scripts/check-awesome-links.mjs@@ -0,0 +1,124 @@+#!/usr/bin/env node+/**+ * Live link check for the Awesome 3D Agents list (data/awesome.json).+ *+ * An awesome list is only worth publishing if every link resolves, so this+ * fetches each url and reports what came back. It is deliberately NOT wired+ * into the deploy path: it talks to a hundred and something third-party hosts,+ * and a flaky CDN must never turn someone else's deploy red. Run it whenever+ * you edit the list.+ *+ * node scripts/check-awesome-links.mjs+ * node scripts/check-awesome-links.mjs --concurrency 16 --timeout 20000+ * node scripts/check-awesome-links.mjs --strict-blocked+ *+ * Three outcomes, because they need three different reactions:+ * ok the url answered. Nothing to do.+ * moved it answered somewhere else. Point the entry at its new home.+ * blocked 403 or 429 from a bot filter. Several of these hosts refuse any+ * request from a datacenter address, so this is a "check by hand",+ * not a dead link. Escalate it with --strict-blocked.+ * broken no answer, or 4xx/5xx that is not a bot filter. Fix or remove.+ *+ * Exits 1 on any broken url, and on duplicate urls, which are the other way an+ * awesome list rots.+ */+import { readFileSync } from 'node:fs';+import path from 'node:path';+import { fileURLToPath } from 'node:url';++const root = path.resolve(path.dirname(fileURLToPath(import.meta.url)), '..');+const argv = process.argv.slice(2);+const flag = (name, fallback) => {+ const i = argv.indexOf(name);+ return i === -1 ? fallback : argv[i + 1];+};++const concurrency = Number(flag('--concurrency', '12'));+const timeoutMs = Number(flag('--timeout', '25000'));+const strictBlocked = argv.includes('--strict-blocked');++const list = JSON.parse(readFileSync(path.join(root, 'data/awesome.json'), 'utf8'));++const targets = [];+for (const section of list.sections) {+ for (const item of section.items) targets.push({ section: section.id, name: item.name, url: item.url });+}++// A duplicate entry is a curation bug the network can never surface, so it is+// checked here rather than in a second script nobody remembers to run.+const seen = new Map();+const duplicates = [];+for (const t of targets) {+ const key = t.url.replace(/\/+$/, '').toLowerCase();+ if (seen.has(key)) duplicates.push(`${t.section}/${t.name} repeats ${seen.get(key)} (${t.url})`);+ else seen.set(key, `${t.section}/${t.name}`);+}++// A browser user agent: several of these hosts answer 403 to anything that+// looks like a script, which would report a healthy page as broken.+const UA =+ 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0 Safari/537.36';++async function probe(url) {+ const options = {+ redirect: 'follow',+ headers: { 'user-agent': UA, accept: 'text/html,application/xhtml+xml,*/*' },+ signal: AbortSignal.timeout(timeoutMs),+ };+ // HEAD first because it is cheap. A host that refuses HEAD, or that only+ // applies its bot filter to HEAD, gets a real GET before we call it broken.+ let head = null;+ try {+ head = await fetch(url, { ...options, method: 'HEAD' });+ if (head.ok) return { status: head.status, finalUrl: head.url };+ } catch {+ head = null;+ }+ const res = await fetch(url, { ...options, method: 'GET' });+ return { status: res.status, finalUrl: res.url };+}++const results = [];+let cursor = 0;++async function worker() {+ while (cursor < targets.length) {+ const target = targets[cursor++];+ try {+ const { status, finalUrl } = await probe(target.url);+ results.push({ ...target, status, finalUrl });+ } catch (error) {+ results.push({ ...target, status: 0, error: error.message });+ }+ }+}++await Promise.all(Array.from({ length: Math.min(concurrency, targets.length) }, worker));++// Compare without the fragment: an anchor into a single-page docs site is+// resolved in the browser, so the server answering the bare page is correct.+const normalise = (u) =>+ u+ .replace(/#.*$/, '')+ .replace(/\/+$/, '')+ .replace(/^https?:\/\/www\./, 'https://');++const blockedCodes = new Set([403, 429]);+const broken = results.filter((r) => r.status === 0 || (r.status >= 400 && !blockedCodes.has(r.status)));+const blocked = results.filter((r) => blockedCodes.has(r.status));+const moved = results.filter(+ (r) => r.status >= 200 && r.status < 400 && r.finalUrl && normalise(r.finalUrl) !== normalise(r.url),+);++for (const r of moved) console.log(`[moved] ${r.name}\n ${r.url}\n -> ${r.finalUrl}`);+for (const r of blocked) console.log(`[blocked] ${r.section}/${r.name} ${r.url} HTTP ${r.status} (bot filter, verify by hand)`);+for (const r of broken) console.error(`[broken] ${r.section}/${r.name} ${r.url} ${r.error ?? `HTTP ${r.status}`}`);+for (const d of duplicates) console.error(`[dupe] ${d}`);++const ok = results.length - broken.length - blocked.length;+console.log(+ `\n[awesome-links] ${ok}/${results.length} reachable, ${moved.length} redirecting, ${blocked.length} bot-filtered, ${broken.length} broken`,+);++if (broken.length || duplicates.length || (strictBlocked && blocked.length)) process.exit(1);diff --git a/tests/awesome-list.test.js b/tests/awesome-list.test.jsnew file mode 100644index 000000000..bd46742df--- /dev/null+++ b/tests/awesome-list.test.js@@ -0,0 +1,112 @@+// The Awesome 3D Agents list (data/awesome.json and its generated outputs).+//+// The list is published in two places at once: awesome/README.md, which people+// read and fork on GitHub, and public/awesome.json, which /awesome renders. A+// generator bug that desynchronises them is invisible until someone notices the+// page and the README disagree, so the invariants that keep both honest are+// pinned here. `npm run check:awesome` already fails the build when the+// committed outputs drift from the source; these tests cover the curation rules+// a regenerate would happily preserve.+//+// Link liveness is NOT tested here on purpose: it needs the network and a+// hundred third-party hosts. That check is `npm run awesome:links`.++import { readFileSync } from 'node:fs';++import { describe, it, expect } from 'vitest';++const source = JSON.parse(readFileSync('data/awesome.json', 'utf8'));+const published = JSON.parse(readFileSync('public/awesome.json', 'utf8'));+const readme = readFileSync('awesome/README.md', 'utf8');++const allItems = source.sections.flatMap((s) => s.items);++describe('Awesome 3D Agents source', () => {+ it('gives every entry a name, an absolute url, and a description', () => {+ for (const item of allItems) {+ expect(item.name, 'entry name').toBeTruthy();+ expect(item.description, `${item.name} description`).toBeTruthy();+ expect(item.url, `${item.name} url`).toMatch(/^https?:\/\//);+ }+ });++ it('links each project once, so the list cannot recommend the same thing twice', () => {+ const seen = new Map();+ for (const item of allItems) {+ const key = item.url.replace(/\/+$/, '').toLowerCase();+ expect(seen.get(key), `${item.name} repeats ${seen.get(key)}`).toBeUndefined();+ seen.set(key, item.name);+ }+ });++ it('keeps descriptions to the one-sentence length the awesome format expects', () => {+ for (const item of allItems) {+ expect(item.description.length, `${item.name} description length`).toBeLessThanOrEqual(260);+ }+ });++ it('honours the repository-wide ban on em-dash and en-dash characters', () => {+ const banned = /[\u2014\u2013]/;+ for (const item of allItems) {+ expect(banned.test(item.name), `${item.name} name`).toBe(false);+ expect(banned.test(item.description), `${item.name} description`).toBe(false);+ }+ for (const section of source.sections) {+ expect(banned.test(section.title + section.description), `${section.id} section`).toBe(false);+ }+ expect(banned.test(readme), 'awesome/README.md').toBe(false);+ });++ it('uses a unique, anchor-safe id for every section', () => {+ const ids = source.sections.map((s) => s.id);+ expect(new Set(ids).size).toBe(ids.length);+ for (const id of ids) expect(id).toMatch(/^[a-z0-9-]+$/);+ });++ it('never ships an empty section', () => {+ for (const section of source.sections) {+ expect(section.items.length, `${section.id} is empty`).toBeGreaterThan(0);+ }+ });+});++describe('Awesome 3D Agents generated outputs', () => {+ it('publishes exactly what the source holds', () => {+ expect(published.counts.items).toBe(allItems.length);+ expect(published.counts.sections).toBe(source.sections.length);+ expect(published.sections.map((s) => s.id)).toEqual(source.sections.map((s) => s.id));+ });++ it('gives every published entry a tags array, so the page can filter without a guard', () => {+ for (const section of published.sections) {+ for (const item of section.items) expect(Array.isArray(item.tags)).toBe(true);+ }+ });++ it('offers only tags that appear more than once as filter chips', () => {+ const counts = new Map();+ for (const item of allItems) {+ for (const tag of item.tags ?? []) counts.set(tag, (counts.get(tag) ?? 0) + 1);+ }+ for (const { id, count } of published.tags) {+ expect(counts.get(id), `${id} count`).toBe(count);+ expect(count, `${id} is offered as a chip`).toBeGreaterThan(1);+ }+ // Most common first, so the filter bar leads with the useful facets.+ const ordered = [...published.tags].sort((a, b) => b.count - a.count || a.id.localeCompare(b.id));+ expect(published.tags).toEqual(ordered);+ });++ it('renders every entry into the README in the awesome-list format', () => {+ for (const item of allItems) {+ expect(readme, `${item.name} in README`).toContain(`- [${item.name}](${item.url}) - `);+ }+ expect(readme).toContain('[](https://awesome.re)');+ });++ it('lists every section in the README contents', () => {+ for (const section of source.sections) {+ expect(readme, `${section.title} in contents`).toContain(`## ${section.title}`);+ }+ });+});diff --git a/vercel.json b/vercel.jsonindex d022e2daf..09b9e17d9 100644--- a/vercel.json+++ b/vercel.json@@ -5650,6 +5650,14 @@"src": "/bundles/","dest": "/bundles.html"},+ {+ "src": "/awesome",+ "dest": "/awesome.html"+ },+ {+ "src": "/awesome/",+ "dest": "/awesome.html"+ },{"src": "/mcp-tools","dest": "/mcp-tools.html"diff --git a/vite.config.js b/vite.config.jsindex 842c8108d..423b58942 100644--- a/vite.config.js+++ b/vite.config.js@@ -589,6 +589,7 @@ const appConfig = {'tour-builder': resolve(__dirname, 'pages/tour-builder.html'),'agent-identities': resolve(__dirname, 'pages/agent-identities.html'),'mcp-tools': resolve(__dirname, 'pages/mcp-tools.html'),+ awesome: resolve(__dirname, 'pages/awesome.html'),'render-lab': resolve(__dirname, 'pages/render-lab.html'),holo: resolve(__dirname, 'pages/holo.html'),bundles: resolve(__dirname, 'pages/bundles.html'),@@ -1344,6 +1345,8 @@ const appConfig = {'/tour-builder/': resolve(root, 'pages/tour-builder.html'),'/agent-identities': resolve(root, 'pages/agent-identities.html'),'/agent-identities/': resolve(root, 'pages/agent-identities.html'),+ '/awesome': resolve(root, 'pages/awesome.html'),+ '/awesome/': resolve(root, 'pages/awesome.html'),// Production routes /app to the Next viewer and leaves app.html// reachable only at /app-classic (vercel.json). Dev has to mirror// that, or every /app change gets exercised against the wrong page.