IAMCCS_goyAIcanvas-easy (legacy compat)

This is a compatibility alias, and once you know that, there's not much else to learn. In the pack's __init__.py, IAMCCS_goyAIcanvas-easy is registered to the exact same class as IAMCCSGoyaCanvasEasyNode - same one-string input, same output, same browser editor, same backend routes. The pack's own web extension lists it as the "legacy" type.

Why does it exist? Because the author renamed the class at some point and didn't want every workflow saved with the old name to break. If you've been using this node since before the rename, your saved graphs keep loading because the alias is still mapped. New graphs should use the canonical name (IAMCCSGoyaCanvasEasyNode) - that's the one the README and the UI treat as primary, and it's where the display label "IAMCCS_goyAIcanvas-easy" actually points.

Nothing about the behavior differs. You still get the same all-in-one canvas editor behind the node logo: text-to-image, image-to-image, draw, inpaint, outpaint, upscale, model selection, LoRA stack, and the gallery that reads ComfyUI/output/goya_output. The goya_easy_state input/output is still just the JSON persistence blob you never edit by hand.

Installation is identical to the main node:

cd ComfyUI/custom_nodes
git clone https://github.com/IAMCCS/IAMCCS_goyAIcanvas-easy

Restart ComfyUI and hard-refresh the browser - the web UI caches aggressively, so that refresh is not optional. No extra dependencies beyond what ships with ComfyUI, no model downloads (the editor lists whatever's in your model folders and extra_model_paths.yaml).

One thing that trips people up: both names may show up when you search the node menu. That's expected, not a bug. Pick the canonical one for anything new, treat this one as a bridge for old workflows, and move on with your day. Same editor, same license (Apache 2.0, attribution to IAMCCS), same caveat that the models you feed it carry their own licenses.