Clothing_State Styler (Advanced)
A wardrobe-state dropdown
- text_positive_g
- text_positive_l
- text_positive
- text_negative_g
- text_negative_l
- text_negative
Clothing_State is the menu for the condition clothing is in, as opposed to what the clothing is. It's a 49-entry dropdown of Danbooru-derived states: unbuttoned, unzipped, untied, open clothes, clothing aside, see-through, undressing. In plain terms, it's the "coming undone" axis - and yes, that makes it one of the pack's NSFW-leaning menus. It exists because these are all real booru tags the SDXL-lineage anime models actually understand, so the menu is a convenient way to reach them without memorizing the vocabulary.
Whether that's useful to you depends entirely on what you're making. If it's not your thing, skip it; if it is, it's a tidy shortcut to a specific set of tags.
The "Advanced" part
This node ships only in the Advanced flavor, built for SDXL's two text encoders (OpenCLIP-G and CLIP-L). You get text_positive_g and text_positive_l instead of one positive box, plus negative_prompt_to (Both / G only / L only). Put the same subject in both encoders to start; the split is optional finesse, not a requirement. It only means anything on SDXL - SD 1.5 has one encoder, and the 2026 LLM-encoded models don't have this CLIP-G/CLIP-L world at all.
How it works
It's a fork of SDXL Prompt Styler. Each state is a JSON template with a {prompt} slot; your text drops in, the state phrase gets wrapped at a strong weight (~2.0), and rewritten prompt strings come out. Text only - nothing loads a model. Because these are literal booru tags, they behave best on the SDXL/Illustrious/Pony family that was trained on that tag vocabulary.
Inputs and outputs that matter
- clothing_state - the 49-entry dropdown.
none= off. - text_positive_g / text_positive_l - your subject, once per SDXL encoder.
- negative_prompt_to -
Both(default),G only, orL only.
log_prompt prints the assembled prompt. The node returns six strings - text_positive_g, text_positive_l, combined text_positive, and the three negatives. Wire _g/_l into an SDXL dual CLIP Text Encode, or take the combined text_positive / text_negative.
Installing it
ComfyUI Manager: search ComfyUI_MileHighStyler, install, restart. Or:
cd ComfyUI/custom_nodes && git clone https://github.com/TripleHeadedMonkey/ComfyUI_MileHighStyler
No models, no dependencies. Nodes appear under ali1234/stylers.
Where people get burned
- Wrong model family. These are booru tags. On an anime SDXL checkpoint (Illustrious, NoobAI, Pony) they land clean; on a photoreal base or a 2026 LLM-encoded model they get vaguer, because the tag vocabulary was never trained in the same way. Match the menu to the model.
- Strong tags conflict. A "state" tag at
2.0can argue with whatever the Clothing_Style menu or your own prompt says the outfit is. If the result is garbled, you've over-specified the wardrobe - thin it out. - The Advanced split is SDXL-only. Off SDXL the
_g/_lsockets are dead weight; use a basic node or just type the tag.
Inputs (6)
| Name | Type | Default | Description |
|---|---|---|---|
| text_positive_g | STRING | — | |
| text_positive_l | STRING | — | |
| text_negative | STRING | — | |
| clothing_state | COMBO | 49 options: none, clothing aside, clothes down, open clothes, revealing clothes, see-through, +43 | |
| negative_prompt_to | COMBO | Both | 3 options: Both, G only, L only |
| log_prompt | BOOLEAN | true | — |
Outputs (6)
| Name | Type | Description |
|---|---|---|
| text_positive_g | STRING | — |
| text_positive_l | STRING | — |
| text_positive | STRING | — |
| text_negative_g | STRING | — |
| text_negative_l | STRING | — |
| text_negative | STRING | — |