UV Select Less

UV Select Less is the shrink-wrap of selections: it peels one layer of UV elements off the edges of whatever is currently selected. Select all → Select Less → and you're left with the interior, minus the boundary ring. It's the complement of Blender's "select more," and in a graph it's how you say "apply to everything except the edges" without listing those edges. Small node, small job - but it's the difference between an unwrap that leaves clean seams and one that chews on the borders of your islands.

How it works

It wraps bpy.ops.uv.select_less on your BPY_OBJ, same auto-generated pattern as the rest of the UV_* nodes: enter edit mode, run the operator, return the object. No inputs beyond the mesh - it just needs a selection to shrink, which comes from a node upstream like UV Select All or UV Select. Output is the same BPY_OBJ with the trimmed selection.

What "one step" means in practice: the operator deselects the outermost selected elements (typically boundary vertices or edge-connected elements depending on the current UV selection mode), leaving the selection one element thinner all around. It's a topology-aware shave, not a random trim.

Where it fits

The honest use in Avatar Graph is defensive editing. Several UV operations behave differently on the boundary of a region - unwrapping edge faces can stretch, rip at the edge can tear into a neighbor island, and marking seams on a full selection can put cuts exactly where you don't want them. Shave first with Select Less, then run the operation, and the border stays untouched. It's also the standard way to get "everything but the outer ring" when you want an interior island to be selected alone. It's rarely the star of a workflow, but it's the kind of node you reach for when an operation keeps misbehaving at the edges.

Installing Avatar Graph

This node ships in the Avatar Graph pack from Avatech. From ComfyUI/custom_nodes:

git clone https://github.com/avatechai/avatar-graph-comfyui
cd avatar-graph-comfyui
python -m pip install -r requirements.txt

then restart ComfyUI with --enable-cors-header. Or ComfyUI Manager → search Avatar Graph → Install.

Standard pack warnings: it pins bpy==3.6.0, which needs Python 3.10.x - anything else fails the pip install. First launch downloads the ~2.5 GB SAM model, so don't read a long start as a hang. And the README flags the pack as still in flux; pin the version once your workflow is stable.