Amage Text
The two-field prompt node that keeps your LoRA triggers out of your prompt
- STRING
Amage Text is the simplest node in the ComfyUI-Amage pack: two text fields, one string out. It's the sort of node that feels pointless until the exact moment it saves you. Its one trick is that the two fields are different - a single-line input_text and a multiline text - which turns out to be a tidy way to keep your LoRA trigger words from getting tangled up in the prompt you keep rewriting.
Why you'd reach for it
When you're iterating on a prompt, the worst part is retyping or re-pasting the trigger phrase that makes your LoRA actually work every single time you change the main prompt. This node splits the difference: input_text holds the trigger word or fixed style tag, text holds the prompt you'll be editing constantly. You swap prompts freely, and the trigger stays put. That's the README's own framing - "a typical use case is adding LoRA trigger words in front of the main prompt" - and it's exactly right.
How it works
There's no mystery to it. The node does one thing: combined_text = f"{input_text} {text}".strip() - concatenate the two fields with a space, trim the edges, return a STRING. That's the entire mechanism. input_text comes first because it's the prefix; the multiline field is the body. Output a single STRING, ready to feed a CLIPTextEncode or anything else that swallows text.
Inputs & output
Two inputs:
input_text- single line, your trigger word or fixed prefix.text- multiline, your actual prompt.
One output: STRING, the concatenation. Wire it into a CLIPTextEncode's text input and you're done.
The honest take
ComfyUI ships a stock Text node, and text-concat nodes are two a penny across the ecosystem - so if you're building a big graph and need a general text-joining utility, this isn't special. Its only value is the layout: the single-line/multiline split is a nudge that keeps triggers visually separate from prompts. It's a workflow-habit helper, not a power tool. If your LoRAs have trigger words and you keep losing them mid-iteration, the nudge is worth it.
Install
It ships in the same small pack as Amage STT, FPS Converter, and All in One - so you get all four whether you wanted them or not.
cd ComfyUI/custom_nodes
git clone https://github.com/Sampkao/ComfyUI-Amage.git
# restart ComfyUI
Or search "Amage" in ComfyUI Manager and install from there. One pack-wide thing to know: there's no requirements.txt, but nodes.py imports transformers at module load (used only by the STT node), so if transformers isn't in your ComfyUI Python environment, this node fails to appear too. pip install transformers into the ComfyUI venv once and the whole pack loads.
Not a node that'll change your life - but if you've ever spent ten minutes wondering why your LoRA stopped working only to realize you dropped the trigger when you pasted a new prompt, you'll get it.
Inputs (2)
| Name | Type | Default | Description |
|---|---|---|---|
| input_text | STRING | — | |
| text | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| STRING | STRING | — |