USA Attire

Every pack has the odd one out. In ComfyUI Globetrotter Nodes - a suite of culturally-specific character prompt generators - the USA Attire node is the one where "traditional dress" means blue jeans and a flannel shirt. Which is honestly fine, and more useful than it sounds: Americana archetypes like the cowboy hat, denim jacket, and baseball cap get butchered by base models as often as any kimono does, and this node makes sure the prompt names them precisely.

What it is

A character-prompt generator from the manifestations/comfyui-globetrotter pack. Pick a person and a wardrobe, get one prose prompt STRING, feed it into CLIP Text Encode as your positive prompt.

How it works

Same engine as the rest of the pack - the node is generated at load time from JSON in data/, it builds the prompt in priority order (subject → appearance → emotion → attire → pose → setting → atmosphere → composition → quality), gender-filters the attire, and expands each garment name into a description. USA data is thin like most of the pack's non-India/Indonesia entries: no regional appearance data, no country-specific poses or landmarks. Four attire categories, all everyday Americana.

The inputs that matter

• Upper_Body - T-Shirt, Flannel Shirt, Denim Jacket. The flannel is the most "iconic American" choice; the denim jacket description keeps it from rendering as a generic hoodie.
• Head - Cowboy Hat, Baseball Cap, Coon-Skin Cap. The cowboy hat is the highest-value option here - models love to cartoon it up.
• Legs - Blue Jeans. Feet - Cowboy Boots, Sneakers.
• gender and age set the subject; the gender filter shapes which garments show in the dropdowns.
• randomization (off/light/moderate/full) and seed - off + fixed seed for consistent characters, full for exploration (full deliberately ignores the seed).
• lora_trigger, custom_prompt, experimental_llm_rewrite - standard tail inputs; the last needs transformers installed and downloads distilgpt2 on first use.

Output

One prompt STRING → CLIP Text Encode (Prompt) → KSampler.

Install

cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/manifestations/comfyui-globetrotter
cd comfyui-globetrotter
pip install -r requirements.txt   # optional

Restart ComfyUI, or search "Globetrotter" in ComfyUI Manager. No model downloads.

Troubleshooting

• This feels thin for a "culture" pack: it is. If you want richer regional American data, you're expected to add it - the pack's data files (data/attire/*/us.json) are designed to be extended, and the README documents the schema.
• Cowboy hat still renders as a cartoon: archetypes are strong in base models; pair it with a photorealistic detail level and a portrait composition.
• Output wanders between runs: randomization defaults to moderate; set it to "off" and keep the seed for reproducibility.
• experimental_llm_rewrite no-ops without transformers; it skips silently.

The USA node is the plain one in the family, but for cowboy boots-and-flannel character prompts it gets the job done with zero hand-writing. Sometimes the "least exotic" node is the one you actually use.