Part of the Lonecat's Drop In Modules Collection.
Please use and share in your workflows. I designed these for the community 🙏
Designed to plug into any workflow:
Instagram: https://www.instagram.com/synth.studio.models/
Buy me a☕ https://ko-fi.com/lonecatone
This represents many hours of work that I give away for free. If you enjoy it, Please make sure to 👍like and post to my workflows. That is how you support me!!! Also, feel free to ⚡tip 😉
What it does:
A plug & play .json and detailed prompt builder to add to existing workflows
也有中文说明
V2.0
Uses LC123 nodes: https://github.com/lonecatone23/ComfyUI_LC123_nodes
🗒️LC Prompt Builder
multiple detailed choices for character, outfits, style, locations
choose a color palette from a preset or use an image
custom wildcard selector that does not require special words that can plug into ANY prompt
all nodes output strings, so you can use them individually or as a whole
🧩LC assembler creates a fully detailed prompt:
Choose a detailed prompt
choose a .json with bboxes for objects and subjects
toggle objects if not needed
Works with Krea2 Regional Builder as well as Ideogram 4 Prompt Builder KJ
V1.0
Builds .JSON prompts
Allows control over subject, camera, color Palette, style, and scene
Cut & Paste ready to drop in your own workflow
Works well with Flux2, Ideogram, Flux2 Pro, and other .JSON prompt compatible models
Description
FAQ
Comments (7)
Hello. this looks really cool but - can we use it with a SDXL workflow, too? I suppose we need to parse the JSON output before feeding it to the CLIP Text Encoder, right? But, how? Thanks.
hmmm,
Yeah, and I can make that. Didn't think of that. Thee I buit this to plug in before Ideogram.
I'll add a parser sometime in the next day or so
@lonecatone23 That would be fantastic - and I also wonder if then we could run both this workflow and Omost in parallel and merge the results using a Conditioning (combine) node to feed into the CLIP Text Encoder. Overkill?
@McBob67 why? There is a group of nodes already that does prompt building for SDXL. This is overkill as it is.
@lonecatone23 Just toying with the idea of having this neat workflow compose the scene and Omost "direct" it and then merging the inputs...
@McBob67 Did you see version 2? It gives you what you asked for
@lonecatone23 Oh you mean the parser? Outputting both json and "normal" prompt? I had actually put my nose to the grind after we commented about it and (with a little help from Gemini) I had come out with a small and very simple parser node myself - version 2 layout is much better though, so I guess I'll use that; it was a nice opportunity to tinker and learn, anyway. Thank you.