EigenAI Text Processor

The EigenAI Text Processor is the least glamorous node in the Eigen AI FLUX API plugin, and it's honest about it. Open one up and you get a big multiline box pre-filled with "A beautiful landscape painting in Studio Ghibli style" - which is a strong hint about the LoRA this pack defaults to. You type, it passes your text along. That's it.

So why does it exist at all? Because the modular half of this pack is wired around custom types. The generator nodes (EigenAIFluxGeneratorNode, EigenAIQwenGeneratorNode, EigenAIKontextGeneratorNode) expect their prompt input to arrive as a PROMPT-typed value, and this node is what produces one. If you'd rather skip the ceremony, every generator also accepts a plain string - but following the pack's own intended flow, you drag one of these in, type, and wire the prompt output into a generator's prompt input.

What it actually does

Under the hood it's a prompt.strip() and a pass-through. It trims leading/trailing whitespace, logs the result, and returns it. Two details matter more than the trim:

• The input is capped at 2000 characters and renders as a textarea, so it doubles as your prompt workspace when a workflow calls for one.
• It declares IS_CHANGED, so the node re-executes whenever the text changes. Wired into a generator that posts to a remote API, that means editing the prompt retriggers the whole generation chain rather than serving you a cached result. Fine for a prompt, worth knowing before you click.

Inputs: one required field, prompt (STRING, multiline). Output: one prompt (PROMPT type). Nothing else. There is no prompt-enhancement or magic rewriting happening here, whatever the docstrings in the source imply - the code literally passes your text through.

Installing

Same story as every node in this pack - it's an API-client plugin, so there's no model download and nothing heavy:

cd ComfyUI/custom_nodes
git clone https://github.com/jialuw0830/flux_api_comfyui_plugin

Then restart ComfyUI. Or use ComfyUI Manager and search "Eigen AI FLUX API" or "flux api". Dependencies are requests, pillow, numpy, torch - all already present in a stock ComfyUI install, plus an optional colorama.

When you'd bother

Honestly: this is the node you reach for to keep a modular workflow tidy, or not at all. If you're using the unified nodes (EigenAIFluxNode and friends), the prompt field is already on them and you never see this. If you're hand-wiring a modular graph, it gives you one obvious place to swap prompts without digging into generator settings. It's a folder, not a feature - but every pack needs its folders.