📝 Polza Show Text

Every pack that talks to an LLM eventually ships a "show me the text" node, and Polza Show Text is ComfyUI-Polza's version. It does exactly one thing: take any STRING, display it directly on the node in the graph, and hand it back out. That's it. It's the least impressive node in the pack and the one you'll reach for most once you're actually chaining LLM calls.

Here's the problem it solves. ComfyUI is a graph editor, and text outputs don't render themselves - wire Polza Chat or Polza Vision into anything and the reply exists only as an invisible value on a wire. Polza Show Text makes the result visible where you're looking: run the graph, and the node grows a read-only text widget showing whatever came back. Want to see whether that json_object response parsed cleanly, or what GPT-4o actually said before you chain it into an image prompt? This is how you look.

How it works

The node itself is a few lines of Python - it's marked as an output node, takes a single text input, and returns text unchanged in both its UI payload and its result. The real work lives in the pack's web/polza_display.js: that script catches the node's ui.text message after execution, injects a read-only, multiline display widget into the node, and auto-sizes the box to fit. So it's a pass-through with a monitor bolted on. Because it echoes the text back out as a STRING output, you can also tap the wire mid-graph - show it in one place, keep feeding it forward in another.

Inputs and outputs

The whole thing is one input and one output:

• text (required) - any STRING: an LLM reply, a filename, a prompt fragment, whatever.
• text (output) - the same string, passed through untouched.

Wire the text output of Polza Chat, Polza Vision, or any other text-producing node straight in. It's listed under the pack's 🤖 Polza.ai/utils category, and it needs no API key, no network, no model - it's the one node in this pack that works offline.

Installing it

You don't install this node alone - it ships with ComfyUI-Polza. Via ComfyUI Manager, search ComfyUI-Polza, install, restart. Or:

cd ComfyUI/custom_nodes
git clone https://github.com/PioneerMNDR/ComfyUI-Polza
cd ComfyUI-Polza
pip install -r requirements.txt

Then restart ComfyUI and look under 🤖 Polza.ai/utils. No model downloads, no extra dependencies beyond what ComfyUI already bundles.

Issues worth knowing

There's not much to break here, but a couple of things catch people. If the display widget doesn't appear after a run, you're usually on a stale browser tab - hard-refresh so the extension JS reloads. And note the display only updates on execution, so if the node sits there empty, it means that branch of the graph hasn't run yet, not that it's broken. One honest limitation: the widget shows plain text, so a multiline LLM reply renders as-is with no markdown or syntax highlighting. That's a feature if you're inspecting raw json_object output - the unformatted truth is usually what you want to see before you trust it.