VRGDG_PromptSplitterForFMML
The scene-picker splitter that defaults to an array
- text_output_1
- text_output_2
- text_output_3
- text_output_4
- text_output_5
- text_output_6
- text_output_7
- text_output_8
- text_output_9
- text_output_10
- text_output_11
- text_output_12
- text_output_13
- text_output_14
- text_output_15
- text_output_16
VRGDG_PromptSplitterForFMML is another member of the pack's index-based splitter family, and apart from two small differences it's a clone of VRGDG_PromptSplitterForFL: JSON in, an index to pick the scene, sixteen text outputs out.
The two differences are worth knowing because they'll affect what you paste in:
- Its
json_stringinput defaults to[]- an empty array - rather than{}(an empty object). The "FMML" variant expects a JSON array of scene entries, one element per index, instead of a keyed object. - It's tied to a different section of the pack's workflow than the FL version. The acronyms aren't documented anywhere in the README, so don't burn time trying to decode them - treat "FL", "FMML", and "Manual" as workflow-internal labels, and use whichever one the workflow you loaded references.
How it works
Two required inputs:
json_string- a JSON array of scene entries as text. Element 0 is scene 0, element 1 is scene 1, and so on.index- which scene to pull, 0 to 10000. Zero-based, so the first scene isindex: 0.
Outputs: text_output_1 through text_output_16, all STRING - the selected scene's entry unpacked into sixteen slots for whatever the workflow section consumes. Which slot means what isn't documented beyond the ports themselves; in practice you wire them into the video-generation chain in the order the workflow expects.
Where it sits
Like its siblings, this is loop plumbing: a stepper advances the index, the splitter pulls that scene's parameters, and the video node renders one chunk per run. The array-based format suggests it reads the output of a node that writes scenes as a flat list rather than a keyed map - the pack's LLM node can emit either shape depending on how it's asked.
Installing it
Part of the pack:
cd ComfyUI/custom_nodes
git clone https://github.com/vrgamegirl19/comfyui-vrgamedevgirl
restart, or ComfyUI Manager → search vrgamedev. No models, no heavy deps.
The honest take
If your prompt map is an object, use the object-based splitter; if it's an array, use this one. That's genuinely the whole decision - the index-picking behavior is otherwise identical. When in doubt, match the variant to the workflow that shipped with it and you won't go wrong.
Inputs (2)
| Name | Type | Default | Description |
|---|---|---|---|
| json_string | STRING | [] | — |
| index | INT | 00–10000 | — |
Outputs (16)
| Name | Type | Description |
|---|---|---|
| text_output_1 | STRING | — |
| text_output_2 | STRING | — |
| text_output_3 | STRING | — |
| text_output_4 | STRING | — |
| text_output_5 | STRING | — |
| text_output_6 | STRING | — |
| text_output_7 | STRING | — |
| text_output_8 | STRING | — |
| text_output_9 | STRING | — |
| text_output_10 | STRING | — |
| text_output_11 | STRING | — |
| text_output_12 | STRING | — |
| text_output_13 | STRING | — |
| text_output_14 | STRING | — |
| text_output_15 | STRING | — |
| text_output_16 | STRING | — |