executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
multimodal-and-spatialRoute scvi-tools tasks involving multimodal, ATAC, spatial, methylation, velocity, perturbation, contrastive, and external specialized model families.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this sub-skill when a user needs scvi-tools guidance for non-basic single-cell models beyond core RNA-only workflows: CITE-seq, RNA+protein, RNA+ATAC, ATAC-only, methylation, spatial mapping/deconvolution, velocity, perturbation/contrastive analysis, or external specialized models.
scvi.model.TOTALVI; for semi-supervised labels with protein, use scvi.external.TOTALANVI.scvi.model.MULTIVI on MuData; for ATAC-only use scvi.model.PEAKVI.scvi.external.Tangram, scvi.external.RNAStereoscope, and scvi.external.SpatialStereoscope; verify reference labels before choosing Stereoscope.scvi.external.METHYLVI or scvi.external.METHYLANVI depending on whether labels are needed.scvi.external.VELOVI, MRVI, RESOLVI, ContrastiveVI, SysVI, DIAGVI, or Decipher only after matching their required data keys.CYTOVI, SCAR, SCBASSET, GIMVI, or Tangram when their input assumptions match; expect optional extras for some workflows.AnnData registration, setup_anndata, count layers, batches, labels, covariates, and MuData basics, route to the data-setup sub-skill first.SCVI, SCANVI, LinearSCVI, AUTOZI, or SOLO, route to the core-models sub-skill.