Prompt Split
Add Node → Prompt → Prompt Split
Takes one block of text and splits it into up to ten separate prompts, one per generator.
What’s on the node
| Area | What goes in it |
|---|---|
| Text preview | Shows the text arriving from upstream. Read-only: Prompt Split has nothing to type into, the text is always written somewhere else and wired in. |
| Separator buttons | Which character marks a break: -, * or +. |
| Output counter | How many chunks the text split into. It updates as the upstream text changes. |
| P | Required, on the left: this is where the text comes from. A Text Area / Log, a Prompt Generator, or a finished result’s own prompt handle. With nothing connected there’s nothing to split. |
| P | The outputs, on the right: up to ten numbered rings, one per chunk, in the order they appear in the text. |
More on handles in How nodes connect.
How it works
Prompt Split doesn’t take typing of its own — the text has to be wired in. Put your prompts in a Text Area / Log, separated by lines containing just a -, * or +, and connect it to the P input. The node splits on those lines and gives you back a numbered output for each chunk — ring 1 carries the first chunk, ring 2 the second, and so on. Wire each ring to its own generator.
It only splits on a line that’s nothing but the separator, so a prompt full of bullet points or hyphenated words passes through untouched.
When to use this instead of the Prompt Generator’s own Output setting
The Prompt Generator can already produce several prompts at once through its own Output control — that’s the quicker route. Reach for Prompt Split instead when you’re working with text you wrote yourself in a Text Area, or when you want to choose your own separator.