I would like to suggest adding support for the following atlases:
CIT168 Reinforcement Learning Atlas: This is highly valuable for high-resolution subcortical segmentation (especially for ventral striatum and midbrain structures) and is a standard output in many modern pipelines. I also suggest adding the amygdala and hippocampus parcels from the HCP CIFTI subcortical parcellation, as done to create the 4SX atlases that are part of the built-in atlases available through QSIRecon, XCP-d, etc.