Text Join Three Pixaroma

Text Join Three Pixaroma is the middle child of the Text Join family, and it's the size most prompt-building actually needs. Joins three pieces of text into one, each on its own line, every line type-or-wire: type straight in, or drag a wire onto its dot to pull text from another node (the wire wins when connected). Fixed text you typed and live text from an LLM node or prompt generator mix freely, and the lines grow as you enlarge the node so long text has room.

What separates it from a plain concatenate node is the join logic. Right-click for the settings and you choose what goes between the pieces: comma, space, new line, none, or your own. Turn on skip empty and an unconnected box never leaks a stray separator into the result - which is exactly the bug that fills prompts with ,, when you're still wiring things up. And you can rename each line to whatever fits your workflow (subject, lighting, quality) so the node reads like a sentence instead of three numbered boxes.

Inputs and output

• text_1, text_2, text_3 (STRING, multiline) - the three pieces to join.
• text output - the three pieces joined with your separator, ready for a CLIPTextEncode or any text consumer.

Per-line copy and paste sits on each box, and it works in both the classic and the new node interface.

Install

It ships with the Pixaroma pack: ComfyUI Manager → search Pixaroma → install → restart, or:

cd ComfyUI/custom_nodes
git clone https://gitlab.com/pixaroma/comfyui-pixaroma.git

No models, no dependencies. The pack-wide gotcha: after an update, hard-refresh the browser tab (Ctrl+Shift+R / Cmd+Shift+R) - the browser caches old node visuals, and a stale-looking node after an upgrade is almost always that.

Three lines is the classic sweet spot: subject + lighting + quality booster, or a wired prompt + fixed trigger words + a style fragment. If you need one more, there's a four-line version; if you only need two, the two-line one is less node to look at.