Prompt Reader Pixaroma

Prompt Reader Pixaroma - load an image generated with ComfyUI (or Automatic1111 / Forge) and read the positive prompt saved inside its PNG metadata. No image preview, just the text. Outputs the prompt as STRING so you can wire it into a CLIPTextEncode or any other text input and re-use it. Drag-drop a PNG onto the node, click Upload Image, or pick from the file combo. The readout updates the moment a file is selected, so you see the prompt before running the workflow. If the image has no embedded prompt (JPG, screenshot, or a PNG that lost its metadata), the readout shows a short explanation and the STRING output carries the same explanation so downstream wiring does not break. Handles ComfyUI workflows with chained text nodes (ConditioningCombine, StringConcatenate, SDXL dual-text encoders) and the Automatic1111 / Forge 'parameters' format. You can also wire a filename into the optional 'filename' input (for example from Load Image Pixaroma's 'filename' output) - while it is connected the node ignores its own picker and reads the prompt from that image instead. Pick, upload, or drop a file to take over and the wire disconnects.