threejs-world-generation

Generate deterministic, editable Three.js worlds with terrain, regions, landmarks, and camera paths.

1|Updated Aug 28, 2026
One-click install
npx skills add https://github.com/mooch10/mochiptos --skill threejs-world-generation-mooch10
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: threejs-world-generation
Source: https://github.com/mooch10/mochiptos/tree/main/frontend/.agents/threejs-world-generation
Command: npx skills add https://github.com/mooch10/mochiptos --skill threejs-world-generation-mooch10

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Building a coherent 3D environment from a text brief usually produces disconnected shots or untextured blockouts. This Skill turns a world brief into a single persistent, editable Three.js scene graph with semantic regions, terrain, landmarks, scatter, and a camera fly-through, all driven by one shared world specification. ## Core Features & Use Cases - World Spec Authoring: Define world scale, seed, semantic regions, atmosphere, landmarks, and camera keys in one normalized JSON contract. - Two Fidelity Tiers: Use blockout for fast layout validation or production tier with licensed GLTF/GLB catalogs, PBR terrain layers, and asset provenance manifests. - Render-Guided Refinement: Run bounded iteration loops with cinematic, semantic, and wireframe render modes plus global, regional, and walk-level snapshot review. - Use Case: Given a brief like "a volcanic rift dividing a living valley at dawn", produce a 60-second deterministic camera fly-through over continuous terrain with stable region and landmark IDs that remain editable after rendering. ## Quick Start Use the threejs-world-generation skill to turn this world brief into a deterministic, editable Three.js world with a cinematic camera sequence.

Frequently Asked Questions about threejs-world-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a 3D world from a text brief with Three.js?▼

Author a world_spec JSON defining world size, seed, semantic regions, atmosphere, landmarks, and a camera path, then call the threejs_world tool with operation build. The tool emits an editable HyperFrames workspace including world.json, world-runtime.js, and a validation report.

What is the difference between blockout and production quality tiers?▼

Blockout uses procedural primitives and vertex colors for fast layout and semantic validation. Production requires licensed local GLTF/GLB catalogs, at least eight models across four semantic categories, three PBR terrain layers, and forbids primitive landmark fallbacks.

When should I use Three.js versus Blender for 3D world rendering?▼

Choose Three.js when the deliverable is a code-native, browser-rendered editable world. Choose Blender for reference-grade hero rendering of dense textured scenery, then package its numbered frames with FFmpeg. Record the choice at proposal and do not switch silently.

Can I refine one region of a generated world without rebuilding everything?▼

Yes. Keep the seed, region IDs, and camera times fixed, change only the narrowest relevant spec fields for the affected region, and rebuild. Stable IDs keep review notes addressable across up to three bounded refinement rounds.

What are the limitations of procedural Three.js world generation?▼

It does not reproduce photoreal generative 3D pipelines, articulated assets, game physics, navigation meshes, or interaction logic. It also forbids unseeded randomness, remote model fetches, and render-time asset downloads to preserve determinism.