VRGDG_MultiStringConcat

VRGDG_MultiStringConcat glues up to 20 multiline text fields together into one string, with a delimiter you choose. In a world where prompt nodes only take one big text box, this is the escape hatch for keeping your prompt in readable, editable chunks and then assembling them at the end.

The mechanism is dead simple: each string_1 through string_20 is a widget, and the node joins every non-empty one with the delimiter, in order. That last bit matters - empty strings are skipped, so you can leave a slot blank without injecting stray blank lines or commas into the middle of your prompt.

Inputs worth setting

You could fiddle with twenty widgets, but really there are three knobs that matter:

• string_count - how many text boxes show up on the node. Set it to the number of chunks you actually use so the canvas stays tidy (range 1–20, default 2).
• delimiter - what goes between chunks. Default is \n\n (blank line between paragraphs), and you can use \n for a single newline, \t for a tab, or clear it for no separator at all.
• string_1 … string_20 - the actual content. Multi-line enabled, so each one can hold a paragraph.

Output is one text string ready to feed into a CLIP text encode, an LLM prompt node, or whatever is next in line.

Why this is genuinely handy in this pack

The VRGameDevGirl music-video workflow builds prompts the way you'd assemble a shot list: character description in one box, environment in another, lighting, camera motion, interactions, each as a separate editable field. MultiStringConcat is where they all come back together into the single string an encoder or an LLM wants. Keeping them separate means you can tweak the lighting without re-reading the whole three-paragraph prompt, and it makes the graph legible to your future self.

The same trick applies outside the pack. If you've got a base prompt you reuse and a per-scene variation, two boxes and a \n delimiter beats editing one long string every run. And because it skips empties, a single node can serve as the "assembly" point for several optional segments without you toggling anything.

Install

It's part of comfyui-vrgamedevgirl, so:

cd ComfyUI/custom_nodes
git clone https://github.com/vrgamegirl19/comfyui-vrgamedevgirl

Restart ComfyUI and look under VRGDG/General. ComfyUI Manager users can search vrgamedev instead. Nothing else to download - the pack's heavier dependencies (kornia, librosa, imageio) are for the video and audio nodes, not this one.

The honest part

This is a utility, not a feature. If you already have a concat node from another pack it does the same thing, and ComfyUI's native String Concatenate covers the two-or-three-input case. The differentiators here are the 20 slots and the skip-empty behavior, which become worth having exactly when you're assembling big prompts from many optional parts - which, in this pack's workflow, you are.