Link Comfy Nodes
A collection of custom ComfyUI nodes by Mister-Link, including color parsing, frame calculation, and utility nodes for enhanced workflow capabilities.
Nodes (40)
Slot one frame into an existing image batch
Four ways to find the subject and crop to it
Measure how big your tracked mask actually is
Auto-numbered frame exports without the filename wrangling
Template-based text building with %1, %2 placeholders
Quantize and block your frames into retro pixel art
Apply one exact crop box across a whole frame batch
A custom dropdown widget, plus the index you actually need
Auto-pick a color that will never blend into your image
A lighter, quicker preview than the stock node
Hex, 24-bit int, and RGB, all from one field
Get your text-to-motion skeleton out of ComfyUI and into Blender, Unity, or Maya
Type 'a person walks forward and turns around', get an actual 3D skeleton doing it
Text-to-3D-motion in ComfyUI starts here — and the first run downloads about a gigabyte
Watch your generated motion as a playable 3D skeleton before you export anything
A before/after slider without leaving your graph
Batch-rotate frames with a real background fill
Both denoise states from one sampler pass
Pull a whole image sequence in as one batch
The color-match node that fixes flicker, because it understands you're working with video
Dial WAN's built-in pose control up, down, or past normal
Snap a frame batch onto a fixed color swatch
The pixel-art node that knows the model didn't make pixel art — it makes this
Real color-quantized pixel art, not a blur-and-shrink trick
Read text output in your graph, formatted properly
Actually see your transparency before you export it
Swap or inject transparency into a batch of frames
Keep your mask perfectly in step while you resize
Archive an output folder without leaving ComfyUI
Archive multiple frame/mask streams straight from your graph
Circularly rotate a frame sequence
Move the seam to the middle and let the model paint over it
Round a number to a multiple your model actually needs
Turn a real-world size into pixel dimensions
Pack a frame batch into one sheet at a target aspect ratio
Play back a spritesheet before you ship it
Turn a comma-separated string into an actual list
The other half of the WAN loop trick — undo the shift once the transition frames exist
Round any frame count to WAN's n×4+1 pattern
Pad or trim a clip toward a target length, WAN-valid
AI slop nodes for my own purposes, use at your own risk (I change how nodes behave without warning).