StyleRef Facets
The style, cracked open section by section
- style
- output_format
- container_boundary
- mood_personality
- colors
- typography
- light_shadow
- spatial_hierarchy
- surface_material
- motion
- shape_language
- stroke_system
- voice_language
- post_processing
- background_environment
- artistic_mediums
- ui_web
- references
- guardrails
- custom_style_item_1
- custom_style_item_2
- custom_style_item_3
- custom_style_item_4
- custom_style_item_5
- custom_style_item_6
Let me be honest up front: most of the time you won't need this node. That's not a knock on it - it's the correct framing. StyleRef Facets takes the style object from StyleRef Load and splits it into one output per schema section, so you can see what a style actually carries and route the pieces wherever you want them. It's a debug and plumbing tool, not a generation node, and it earns its keep in exactly two situations.
Situation one: recon. Queue the Facets template with a style you're considering, and you can read what's actually in it - the lighting direction, the palette, the medium, the guardrails. This matters more than it sounds, because "the style reads generic" is a real failure mode on long styles: StyleRef Apply sends every section by default, and the identity-carrying ones render last. Before you start guessing which sections to keep on Apply, look at the facets. It turns "tune blindly" into "tune informed."
Situation two: routing. Every section comes out as its own string, so you can wire colors to one place, lighting to another, the mood text somewhere else. If you're building a graph that treats a style's parts as independent ingredients - say, feeding one facet into a different encoder than the rest - this is the node that makes that possible.
How it works
Mechanically it's the thinnest node in the pack: it reads fields off the style dict that Load already fetched and hands them out. No network call, no compilation, nothing to configure. Wire style in, get strings out.
Inputs and outputs
One input, style (from StyleRef Load). The outputs are the style schema's full section list, exactly - output_format, container_boundary, mood_personality, colors, typography, light_shadow, spatial_hierarchy, surface_material, motion, shape_language, stroke_system, voice_language, post_processing, background_environment, artistic_mediums, ui_web, references, and guardrails - plus six custom_style_item_1 through custom_style_item_6 for the six custom items a style can define. Everything is a readable-lines string.
An empty output usually means the style simply doesn't carry that section - a minimalist style has no typography and nothing's broken. The custom items are empty when unset. If every output is empty, you've probably wired something that isn't a StyleRef style, so check the Load node.
Install
Same as the rest of the pack: ComfyUI Manager (search StyleRef) or
cd ComfyUI/custom_nodes
git clone https://github.com/StyleRef/styleref-comfyui
Restart, look for [StyleRef] v1.0.6 - 5 nodes registered in the console. No Python dependencies - the pack runs on the standard library alone. The one real requirement is that styleref.io is reachable when Load first fetches a style; Facets itself does no networking after that. Nothing in the pack spends credits.
Inputs (1)
| Name | Type | Default | Description |
|---|---|---|---|
| style | STYLEREF_STYLE | From StyleRef Load. |
Outputs (24)
| Name | Type | Description |
|---|---|---|
| output_format | STRING | The style's output format fields as readable lines. |
| container_boundary | STRING | The style's container boundary fields as readable lines. |
| mood_personality | STRING | The style's mood personality fields as readable lines. |
| colors | STRING | The style's colors fields as readable lines. |
| typography | STRING | The style's typography fields as readable lines. |
| light_shadow | STRING | The style's light shadow fields as readable lines. |
| spatial_hierarchy | STRING | The style's spatial hierarchy fields as readable lines. |
| surface_material | STRING | The style's surface material fields as readable lines. |
| motion | STRING | The style's motion fields as readable lines. |
| shape_language | STRING | The style's shape language fields as readable lines. |
| stroke_system | STRING | The style's stroke system fields as readable lines. |
| voice_language | STRING | The style's voice language fields as readable lines. |
| post_processing | STRING | The style's post processing fields as readable lines. |
| background_environment | STRING | The style's background environment fields as readable lines. |
| artistic_mediums | STRING | The style's artistic mediums fields as readable lines. |
| ui_web | STRING | The style's ui web fields as readable lines. |
| references | STRING | The style's references fields as readable lines. |
| guardrails | STRING | The style's guardrails fields as readable lines. |
| custom_style_item_1 | STRING | Custom style item 1 — empty when unset. |
| custom_style_item_2 | STRING | Custom style item 2 — empty when unset. |
| custom_style_item_3 | STRING | Custom style item 3 — empty when unset. |
| custom_style_item_4 | STRING | Custom style item 4 — empty when unset. |
| custom_style_item_5 | STRING | Custom style item 5 — empty when unset. |
| custom_style_item_6 | STRING | Custom style item 6 — empty when unset. |