- Duplicate: derive the shift delta from the copy's actual bounding box (not the group frame, since a member can protrude after individual editing) and apply one uniform delta with no per-clone clamp, so the copy's internal layout is preserved exactly. Falls back to a negative shift when the group is pinned to an edge (no more copy-on-top-of-original). - Mirror: skip whole LOCKED sub-branches as a unit (collect members without crossing a locked box) so a nested sub-group frame and its children always mirror together; clamp the whole reflected interval (keeps box size, stays on-canvas) instead of one edge; validate the axis. JS parses clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ComfyUI-AI2Go-Utils
ComfyUI nodepack by AIKnowledge2Go
Contact
- YouTube: @AIKnowledge2Go
- Patreon: AIKnowledgeCentral
- Book an appointment: 1hr 1-on-1 Stable Diffusion
- Email: BeyondMatrixDevelopments@gmail.com
- Civitai: AIknowlege2go
Installation
Clone into your ComfyUI custom_nodes directory and restart ComfyUI:
cd ComfyUI/custom_nodes
git clone https://github.com/Little-God1983/ComfyUI-AI2Go-Utils
Nodes
AI2Go Ideogram 4 Prompt Builder
A visual prompt builder for Ideogram 4's structured JSON caption format. Draw bounding-box regions on a canvas, set each region's type / description / text / colors, and the node assembles the caption JSON.
Includes an Overview panel — a docked, collapsible tree of every region. Click a row to select it (two-way with the canvas), so you can pick and edit regions without accidentally nudging them on the canvas. Ctrl/Cmd-click to multi-select, Shift-click for a range; lock, duplicate, delete, and drag a row onto another to nest it (parenting — move a parent on the canvas and its children follow). The hierarchy is editor-only; the exported caption stays flat.
License & Credits
Licensed under GPL-3.0 (see LICENSE).
This pack builds on the Ideogram 4 Prompt Builder node and editor from ComfyUI-KJNodes by Kijai, which is licensed under GPL-3.0. Derived files retain their original attribution.