Halloween Styler (Advanced)

The Halloween Styler's 51 horror presets - Haunted Asylum, Voodoo Priestess, Witch's Spellbook - but in the Advanced flavor: two positive boxes, six outputs, and a negative_prompt_to dropdown. HalloweenStylerAdvanced from aegis72/comfyui-styles-all is built for SDXL workflows that want to feed a global prompt to the OpenCLIP-G encoder and a local one to CLIP-L.

The inputs

• text_positive_g / text_positive_l - your two positive boxes.
• text_negative - your negative; the style's is appended.
• Halloween - the 51-option dropdown.
• negative_prompt_to - Both (default), G only, or L only.
• log_prompt - on by default; prints the before/after to the console.

Outputs: text_positive_g, text_positive_l, text_positive (combined), text_negative_g, text_negative_l, text_negative.

The catch with these templates

Same story as the other Advanced nodes in this pack, verified against the code. The Halloween templates are single-part wrappers - no . separator - so the G/L machinery has nothing to split. And the two positive boxes trade places: your text_positive_l text is what gets styled and exits as text_positive_g, while text_positive_g passes through unstyled to text_positive_l.

Path of least resistance:

• Type your prompt into text_positive_l and take the combined text_positive output into CLIPTextEncode (or CLIPTextEncodeSDXL on SDXL).
• If you only want one dropdown and one box, the plain Halloween Styler is the simpler pick - same 51 presets.

Installing it

ComfyUI Manager (search ComfyUI-styles-all), or:

cd ComfyUI/custom_nodes
git clone https://github.com/aegis72/comfyui-styles-all

Then restart ComfyUI. No dependencies, no models.

Where people get burned

• The G/L swap - if your styled output is missing the subject, that's the two boxes trading roles.
• Restart after editing data/Halloween/Halloween.json for new menu options.
• SDXL-era styling - strong on SDXL-lineage checkpoints, weak on Flux and LLM-encoded models, like the rest of the pack.