KitNelo · All tools · Blog

scvi tools

Actual Agent / Skill files

This complete upstream plugin source package includes this skill, other skills, scripts, references, and bundled connection configuration. Preview files show the current skill and essential package information. External connectors, models, runtime dependencies, and accounts still require official setup; individual workflows are not execution-tested.

Deep learning for single-cell analysis using scvi-tools. This skill should be used when users need (1) data integration and batch correction with scVI/scANVI, (2) ATAC-seq analysis with PeakVI, (3) CITE-seq multi-modal analysis with totalVI, (4) multiome RNA+ATAC analysis with MultiVI, (5) spatial transcriptomics deconvolution with DestVI, (6) label transfer and reference mapping with scANVI/scArches, (7) RNA velocity with veloVI, or (8) any deep learning-based single-cell method. Triggers include mentions of scVI, scANVI, totalVI, PeakVI, MultiVI, DestVI, veloVI, sysVI, scArches, variational autoencoder, VAE, batch correction, data integration, multi-modal, CITE-seq, multiome, reference mapping, latent space.

Imported source metadata: runtime behavior, costs and compatibility have not been independently verified. Open the official source for current setup instructions.

Requirements

  • A compatible plugin or Agent Skills client is required; capabilities vary by client.
  • If connecting business data, check the plugin connectors, account permissions and organization policies first.

Use steps

  1. Open the source documentation and confirm scope, supported clients and prerequisites.
  2. Find the bio-research plugin in Claude's plugin browser, or follow the repository's Claude Code instructions to install the complete bundle.
  3. Authorize required connections as instructed by the client, start a new session, try an example task and review the output before delivery.

Example

Example prompt: Use the scvi-tools workflow with my supplied material. Identify missing information first, then produce a result I can review.

Limitations

  • Imported from source metadata; not execution-tested in every client.
  • Copying only SKILL.md can omit connectors, commands, scripts and references; use the publisher’s complete package.
Browse the resource directory