Passthru Input Is List Ovum

Half of ComfyUI debugging is "what the hell is this socket actually carrying?" PassthruInputIsListOvum answers that by doing nothing to your data and telling you everything about it. It accepts any value, interprets it as a list (INPUT_IS_LIST), and dumps a compact pretty-printed version of the value to both your console and the node's canvas status - then passes the value through as a plain Python list. It's a spy node: no transformation, maximum visibility.

How it works

The input is any_in (* - accepts anything). Because the node declares INPUT_IS_LIST, ComfyUI hands it the full list as received rather than a single element, and the node's status output shows you a nicely formatted rendering of the actual value with its type. The any_out output is * typed and returns the value as a plain Python list (this variant does not set OUTPUT_IS_LIST, so downstream sees the raw list object). It also logs the same rendering to the ComfyUI console, with the node's title and ID so you can find it in a busy log.

A deliberate detail: the node's IS_CHANGED always returns NaN, which forces ComfyUI to re-execute it every run. That's exactly what you want from a debug node - it never gets skipped as "unchanged," so the dump is always current.

Why you'd reach for it

When a node downstream keeps erroring on a list that shouldn't be a list (or vice versa), drop this in the wire and read the status. It settles the "is it a list of strings or a list of lists?" argument in one glance. It's also the type-adapter play: this variant forces the value out as a plain Python list - the pack's sibling nodes cover the other combinations (Input Is List + Output Is List, Output Is List only, and a plain passthrough).

Installing it

Ships in comfy-ovum. ComfyUI Manager → search "comfy-ovum", or:

cd ComfyUI/custom_nodes
git clone https://github.com/sfinktah/comfy-ovum

Restart, find it under ovum/debug (display name "Passthru Input Is List Ovum"). No models, no special dependencies.

The gotcha

It's a debugging instrument, not a workflow keystone. Because it always re-executes and logs to console, leaving it permanently in a production workflow adds noise - use it to diagnose, then take it out or mute it. And the pretty-printer collapses long/nested values (you'll see ... for big collections), so a giant tensor dump is a summary, not the full thing. If you need to see the raw value unfiltered, that's what the console log is for.