Prompt Styler Gothic Revival

SDXLPromptbyGothicRevival is the Gothic Revival node in wolfden's SDXL Prompt Styler pack. It carries ~31 presets built around 19th-century Gothic architecture and atmosphere - Gothic Cathedral Splendor, Gothic Haunted Manor, Gothic Revival Castle, Gothic Revival Cemetery, Gothic Revival Lighthouse, and more. Type a subject, pick a scene, and the node wraps your text in pointed arches, tracery, and moody stonework.

Worth being clear on what this is: Gothic Revival is the ornate architectural style - spires, vaulted interiors, dramatic masonry - not the black-eyeliner subculture. If you want haunted-cathedral grandeur, this is your node. (If you're after modern goth-culture or straight horror, the pack's Horror node is a better fit.)

How it works

Each style is a JSON entry with a {prompt} placeholder in its prompt field and a negative_prompt. Pick "Gothic Cathedral Splendor" and your text_positive slots into a template full of soaring arches, stained glass, and dramatic light. No model, no API, no key - string substitution against a lookup table. The architecture comes from the keywords the preset injects around your subject.

The negative is baked into each preset, and anything you type into text_negative gets merged with the template's rather than replacing it.

The inputs that matter

• text_positive - your subject, kept simple. The preset supplies the Gothic setting.
• style - the scene, ~31 options.
• text_negative - optional, merged with the preset's negative.

log_prompt (No/Yes) prints the finished prompt to the console. Outputs are positive_prompt_text_g and negative_prompt_text_g, both strings; positive into the CLIP Text Encode feeding your positive conditioning, negative into the negative one. On SDXL the same string can go into both G and L of a CLIPTextEncodeSDXL.

Where people get tripped up

Architecture stresses the parts of SDXL that struggle with straight lines, symmetry, and repeated structure - expect the occasional bent arch or window that doesn't quite tile. Higher resolution and an upscale pass help; if you need architecturally clean results, a depth or line ControlNet from a reference beats prompting the structure from words.

Because each preset is a whole scene, your subject can get swallowed by the setting. If the mix is off, flip log_prompt to Yes, read what got built, and trim. The styles are JSON in the pack folder.

Installing it

ComfyUI Manager: search SDXL Prompt Styler (wolfden), install, restart. Or manually:

cd ComfyUI/custom_nodes
git clone https://github.com/wolfden/ComfyUi_PromptStylers

Restart ComfyUI and find it under the Style Prompts submenu. No downloads, no dependencies.

Pack-wide caveat: these templates were tuned for SDXL and its CLIP encoder, so they're strongest on SDXL, Illustrious, Pony, NoobAI. On 2026 LLM-encoded models the styling weakens and the inherited negative does nothing at CFG 1, where distilled checkpoints run, because the negative pass isn't computed. If the node doesn't appear after install, restart fully and check Style Prompts, not the root menu.