♻🔥📝 List Looper (Text Generator)

Bjornulf's pack ships a whole cluster of "Text Generator" nodes - dropdowns of pre-written prompt fragments for camera angles, shot types, lighting, poses, and so on, meant to save you from typing "three-quarter view, medium shot, dramatic lighting" from scratch every time. List Looper is the version of that idea built for batch generation instead of single picks: point it at a category, tell it to run through every option in that category, and it queues one generation per value automatically.

How it works

Under the hood this is the same trick every "loop" node in this pack uses: when you set a category to its "ALL" option, the node emits a Python list instead of a single string, and ComfyUI's own execution engine re-runs everything downstream of it once per item in that list. You don't write a loop - you just tell ComfyUI there's a list to consume, and it handles the repetition. That's genuinely how ComfyUI's "looping" works everywhere, not just in this pack; List Looper is just a convenient, pre-populated source of things to loop over.

Inputs and outputs

SELECTION is the category picker: camera_angle, shot_type, lighting, or multi_char_action. Whichever you pick becomes "live" - its own dropdown then offers either a specific value or an ALL (N) option that triggers the loop. camera_angle has 13 real choices (front view, side view, dutch angle, bird's eye view, and so on), shot_type has 10 (close-up through establishing shot), and multi_char_action is the big one at 47 choices - things like "posing for the camera," "dodging a punch," "passing food at dinner." Only the category you selected in SELECTION actually drives the output; the other dropdowns sit there unused for that run.

The single output is a list of text fragments (type *, so it'll connect into most text-accepting inputs) - wire it into a Combine Texts node alongside your actual subject/style prompt, or straight into a CLIP Text Encode if you're feeling minimal.

Installing it

ComfyUI Manager → search "Bjornulf_custom_nodes" → install → restart. Or manually: cd ComfyUI/custom_nodes && git clone https://github.com/justUmen/Bjornulf_custom_nodes, then pip install -r requirements.txt inside that folder before restarting. This node doesn't need anything beyond the pack's base install - no extra model, no extra dependency. It's pure Python string data bundled with the pack.

Common issues & troubleshooting

Budget your GPU time before you hit "ALL" on multi_char_action - 47 choices means 47 full generations queued from one click, and there's no built-in cap or preview-first mode. If you only wanted to spot-check a few variants, pick a specific value instead of ALL and swap it manually.

The category word lists themselves are hardcoded into the pack, not exposed anywhere in the UI for editing. If you want a custom set of angles or actions that aren't on the list, this node isn't the tool - reach for Bjornulf's Advanced Write Text node instead, which supports its own {option_a|option_b|option_c} random-pick syntax on text you write yourself.

Because this pack moves fast - the README's own warning is "very active development, work in progress" - the exact option counts and wording in these categories can shift between versions. Don't be surprised if a future update adds or renames entries in the dropdowns above.