Threejs Skill Router logo

Threejs Skill Router

Community
scottstts
threejs-skill-router

Route ambitious Three.js graphics work to the smallest expert skill set. Use for new visual experiences, graphics rewrites, reference matching, or requests spanning geometry, materials, atmosphere, shadows, temporal effects, and final image treatment.

Overview

Publisherscottstts
RepositoryThreejs-Awesome-Graphics-Agent-Skills
Skill namethreejs-skill-router
Stars
894
Forks
98
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by scottstts on GitHub. Read the source before you install it.

Installation

Install the Threejs Skill Router AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/scottstts/Threejs-Awesome-Graphics-Agent-Skills/tree/main/skills/threejs-skill-router
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/scottstts/Threejs-Awesome-Graphics-Agent-Skills.git /tmp/Threejs-Awesome-Graphics-Agent-Skills
mkdir -p .claude/skills
cp -r /tmp/Threejs-Awesome-Graphics-Agent-Skills/skills/threejs-skill-router .claude/skills/threejs-skill-router
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Threejs Skill Router in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Threejs Skill Router on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Threejs Skill Router is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Three.js Visual Skill Router

Treat the model's Three.js knowledge and official documentation as prerequisites. Load only the expertise that changes the visual result.

This pack goes beyond procedural guidance: its working examples and supporting assets contain much of its practical implementation value. When they are relevant, give them particular attention.

Route by the visual system being authored

Required resultLoad
shot composition, chase/side/orbit rigs, camera handoffs, projection ownership, pointer look, floating origins$threejs-camera-direction
launch and docking timelines, procedural transform phases, springs, staging, rotating-frame alignment, debris motion$threejs-procedural-animation
reusable scalar/vector fields, domain warping, causal masks, procedural normals$threejs-procedural-fields
atlas-filtered blocks, planetary surfaces, hybrid texture-backed PBR soil/moss with procedural displacement and masks, ground and model moss accumulation, terrain wetness, deformable granular sand with mass-coupled grain shading, woven cloth with GPU XPBD and yarn microstructure, simulated fur with interactive grooming and strand shading, lava/emissive surfaces, reflective wave-optical diffraction gratings, thin-film soap bubbles with Airy interference, raytraced diamond/gem refraction, dispersive glass transmission with internal reflection and volume absorption, deforming softbody jelly materials with XPBD mechanics and refractive caustics, authored frame PBR, specular AA$threejs-procedural-materials
height-field ray marching, silhouette-aware POM, curved relief shells, relief self-shadowing$threejs-parallax-occlusion-mapping
well-crafted and complete hard-surface object assemblies, procedural vehicles and humanoid robots, parameter-curve section tracks, tilted shell and spine lofts, pillow panels, exact polygon cuts, UV-owned apertures, spanwise airfoil lofts, sculpted rails/frames, branch rings, fin lofts, profile extrusion, inset, revolve, sweep, solidify, bevels and fillets, shell thickness, direct-topology apertures, semantic mesh writers, material slots and groups, primitive-built forms, coplanar flicker, loose/non-manifold geometry, detached parts, interpenetration, support, clearance and swept-envelope defects$threejs-procedural-geometry
trees, surface-following ivy, painted vines, stylized grass, GPU-computed grass, GPU-culled virtual flower fields, distance-tiered plant geometry, roots, foliage, rooted wind deformation$threejs-procedural-vegetation
buildings, façade grammars, profiles, ornaments, modular mesh writers$threejs-procedural-architecture
planets, terrain, craters, biome fields, coastlines, spherical detail$threejs-procedural-planets
sky scattering, planetary shells, depth-based aerial perspective$threejs-atmosphere-aerial-perspective
weather-driven raymarched clouds and cloud shadows$threejs-volumetric-clouds
FFT oceans, coastal breaker transitions, finite-volume shallow-water beach waves, signed-distance coastlines, shallow-water swash chains, wet sand, rock-impact spray, stylized above/below ocean optics, submerged Snell windows, total internal reflection, forward-refracted structures, pixel-footprint spectral LOD, aquatic perspective, caustic god rays, spectral cascades, choppy derivatives, Jacobian whitecaps$threejs-spectral-ocean
authored analytic waves, bounded heightfield pools, object ripples, differential-area caustics, ray-traced pool volume optics, shared normals, heuristic refraction, fallback absorption, crest foam$threejs-water-optics
falling snow, snow accumulation, model snow caps, wet asphalt puddles, procedural ripple normals, splash flipbooks, rain streaks, shared weather envelopes, surface wetness$threejs-precipitation-surfaces
curved-ray black holes, accretion disks, wormholes and throat transits, null-geodesic integration, lensed celestial spheres and star fields$threejs-raymarched-space-effects
filmic HDR lens-flare compositors, highlight-derived radial ghosts, field-angle pupil deformation, spectral pupil rings, localized glare and bloom, raymarched aurora curtains, finite-footprint emissive slabs, uniform volume integration, equirectangular radiance probes, WebGPU voxel fire and smoke, volumetric fluid fields, mesh-surface emitters, SDF fire collisions, particles, trails, plasma, shockwaves, holographic projections, Fresnel rim shells, scanline banding, layered event effects$threejs-procedural-vfx
accumulated screen frost, touch clearing, wet-window rain, view-aligned droplet refraction and blur$threejs-temporal-surfaces
stable large-world shadows, cascades, clipmaps, cached updates$threejs-shadow-systems
GTAO, bent normals, bilateral reconstruction$threejs-screen-space-ambient-occlusion
HDR bloom and selective emission contribution$threejs-bloom
eye adaptation, tone mapping, LUT grading, output color$threejs-exposure-color-grading
shared depth/normal/velocity ownership and multi-pass ordering$threejs-image-pipeline
fixed-view diagnostics, seed sweeps, temporal and budget evidence$threejs-visual-validation

Execution order

For a new procedural scene:

  1. Define a visual contract: subject, scale, camera distance, motion, and target frame budget.
  2. Load $threejs-camera-direction when framing, lens, camera frame, or mode transitions affect the target.
  3. Load the subject-generation skill.
  4. Add $threejs-procedural-animation when object motion requires authored phases, moving frames, or spring convergence.
  5. Add $threejs-procedural-fields when multiple visual channels must share coherent structure.
  6. Add lighting/shadows and atmosphere only after silhouette and material masks read without effects.
  7. Add $threejs-image-pipeline last.
  8. Load only the atomic image effects actually needed.
  9. Use $threejs-visual-validation for a deterministic evidence set.

Routing constraints

  • Do not load a skill for API setup alone. Inspect the installed Three.js version and use official docs.
  • Do not route “make it beautiful” directly to post-processing. Find the missing authored system.
  • Prefer one strong, inspectable visual rule over several independent noise layers.
  • When adapting a supplied reference, preserve the mechanism that creates its character. Do not reduce it to a generic effect category.
  • Keep object-space, world-space, and screen-space systems separate unless the composition explicitly requires coupling.
  • If no retained skill matches, state that the pack lacks expert coverage for that system. Do not stretch the nearest skill to cover it.

Acceptance gate

A routed task is incomplete until the implementation exposes:

  • deterministic seed or reproducible inputs;
  • visual debug modes for its controlling fields;
  • parameters grouped by perceptual role;
  • an intentional mechanism-backed quality or resolution tier when the system defines one;
  • a no-post baseline that still reads.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Threejs Skill Router AI skill do?

Route ambitious Three.js graphics work to the smallest expert skill set. Use for new visual experiences, graphics rewrites, reference matching, or requests spanning geometry, materials, atmosphere, shadows, temporal effects, and final image treatment.

Why use Threejs Skill Router on TypingMind?

Because you install it once and use it with any model. Threejs Skill Router is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Threejs Skill Router in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/scottstts/Threejs-Awesome-Graphics-Agent-Skills/tree/main/skills/threejs-skill-router. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Threejs Skill Router?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Threejs Skill Router?

As many as you like. As long as a model supports skills, you can use Threejs Skill Router with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Threejs Skill Router AI skill free?

Yes. It is published on GitHub by scottstts under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇