Text Append/Prepend
Glue two strings together, either order
- STRING
Sticks two pieces of text together. That's the node. But it's the kind of dumb-simple building block that ends up in half your prompt-building graphs, because "combine a fixed part with a variable part" is something you do constantly and core ComfyUI doesn't give you a clean node for.
Why you'd reach for it
Prompts are rarely one solid block. You've got a style boilerplate you always want - masterpiece, best quality, sharp focus, cinematic lighting - and a subject that changes every run. Or a quality tag you tack onto the end of whatever a wildcard node spat out. Rather than editing one giant text box every time, you keep the parts separate and join them here. Change the subject, the boilerplate stays put.
That separation is the whole value. It's the difference between a prompt you edit and a prompt you assemble, and assembled prompts are what let you swap one piece without disturbing the rest.
How it works
Three inputs:
- current_text - the base string (multiline).
- new_text - the piece you're adding (multiline).
- mode -
append(new_text goes after) orprepend(new_text goes before).
One STRING output, the joined result. append gives you current_text + new_text; prepend flips it. Wire the output straight into a CLIP Text Encode, or into another append node to keep stacking.
Building bigger prompts
The move is to chain them. Base style → append subject → append quality tags → append LoRA trigger words. Each link adds one controllable piece, and any of them can come from a wildcard, a CSV reader, or a fixed text box. Prepend is there for the cases where order matters - some models weight the front of the prompt more heavily, so putting your subject first with prepend can nudge emphasis without you rewriting anything.
One small thing to mind: this node joins strings exactly as given, so if you want a comma and a space between the two pieces, put it there yourself - either at the end of current_text or the start of new_text. Otherwise you'll get blue skyphotorealistic mashed together, which the encoder reads as one weird token.
Installing ComfyUI-Addoor
ComfyUI Manager: Install Custom Nodes → search "ComfyUI-Addoor" → install → restart. Manual:
cd ComfyUI/custom_nodes
git clone https://github.com/Eagle-CN/ComfyUI-Addoor.git
cd ComfyUI-Addoor
pip install -r requirements.txt
Restart; it's under 🌻 Addoor / text. Nothing to configure, nothing to break - just remember the spacing/comma is on you.
Inputs (3)
| Name | Type | Default | Description |
|---|---|---|---|
| current_text | STRING | — | |
| new_text | STRING | — | |
| mode | COMBO | 2 options: append, prepend |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| STRING | STRING | — |