Create visual system
create_visual_systemTransforms a natural-language description of a visual into a complete, verified node network with preview. Classifies intent and delegates to the appropriate builder.
Instructions
Create a complete visual system from a natural-language description. Classifies intent (audio-reactive, particle, feedback, reaction-diffusion, landscape, generative) and delegates to the matching Layer-1 builder (or a tag-matched recipe), creating a self-contained COMP under parent_path, then verifies and previews it. Use plan_visual instead for a dry run that reports which tool/recipe would be chosen without building anything. Returns a note on how the description was interpreted plus the underlying builder's result (created nodes, exposed controls, and a preview image).
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| description | Yes | Natural-language description of the visual system. | |
| parent_path | No | Parent COMP path the generated visual-system container is created inside. | /project1 |
| resolution | No | Advisory target resolution. Recorded in the build note; the sub-builders use their own internal sizes and do not enforce this per-node. | 1080p |
| target_fps | No | Advisory target frame rate (informational only — TD's real cook rate is a project-level setting, not set here). |