Show Text (Lackluster)
See any string in your graph, right on the node
- STRING
Prompts get assembled from five places, and by the time they hit the encoder you often have no idea what the actual string was. Show Text (Lackluster) is the fix: feed it any STRING and it displays the value directly on the node's widget - a live preview panel for whatever text is flowing through your graph at that moment.
How it works
It's a text input with forceInput, meaning it expects a wire rather than letting you type - this node exists to display, not to author. Pass it a prompt, a label, a filename, an error string from some tool, anything, and it shows the value on the node and passes it straight back out. The output is the same STRING you put in, so you can chain it inline without breaking a connection.
Two implementation details make it genuinely useful:
- It's a list-friendly output node. The output is marked as a list (
OUTPUT_IS_LIST), so it handles batches of strings - a list of per-shot prompts, for instance - and displays them. You're not restricted to single values. - It writes the value back into the workflow. Under the hood it uses the workflow's
extra_pnginfoto store the displayed value on the node, which is why saved workflows show you what text was there. It's a small thing, but it's why a saved workflow stays readable months later.
Where it shines
- Debug a camera node or prompt enhancer: wire
promptout of it into Show Text and see exactly what's being sent to the encoder. - Check a tool that returns a STRING you're not sure about - an outdated-packages list, a settings dump, a loader's filename.
- Verify a list before you kick off a long batch.
Because it's a display node, it also forces execution of whatever's upstream on each run - which is generally what you want from a debug tool.
Notes
- Install with the pack: ComfyUI Manager → "ComfyUI Lackluster Nodes" (
requests+numpy). - Don't confuse it with Text Multiline (Lackluster), which is the authoring box where you type. This one takes a wire and shows you what arrived. Put both in a graph and you've got a complete prompt read/write station.
For a node that does almost nothing, it's one of the pack's most-used in practice - every workflow debugging session ends with a Show Text somewhere.
Inputs (1)
| Name | Type | Default | Description |
|---|---|---|---|
| text | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| STRING | STRING | — |