What problem does it solve?
Creating coherent 3D environments for video production usually requires manual 3D modeling or produces disconnected 2D shots. This Skill builds persistent, editable Three.js scene graphs from text briefs, keeping terrain, regions, landmarks, and camera paths consistent across the whole world.
Core Features & Use Cases
- Deterministic world building: Author a single world_spec with seed, semantic regions, terrain operators, atmosphere, landmarks, and camera paths, then rebuild identically at any time.
- Two fidelity tiers: Use blockout for fast layout validation or production tier with licensed GLTF/GLB asset catalogs, PBR terrain layers, and provenance manifests.
- Render-guided refinement: Inspect global, regional, and walk-level snapshots in cinematic, semantic, or wireframe modes and fix only the affected region while preserving seeds and IDs.
- Use Case: Given a brief like "a volcanic rift dividing a living valley at dawn," produce a 60-second camera fly-through as an editable HyperFrames workspace, then hand off to Blender for reference-grade rendering.
Quick Start
Ask the agent to turn your world brief into a deterministic, editable Three.js world with a camera fly-through using the threejs-world-generation skill.