Lamindb

If you're wrangling biological datasets like scRNA-seq or spatial transcriptomics and need more than a file dump, this gives you queryable data with automatic lineage tracking. It's built around making data FAIR compliant without the usual pain, letting you annotate with proper biological ontologies (genes, cell types, diseases via Bionty) and validate schemas before things break downstream. The workflow integration is solid, hooks into Nextflow, Snakemake, Weights & Biases, and MLflow so your computational provenance actually gets captured. It's Python-based lakehouse architecture that tracks what notebook produced what dataset from what inputs. Best for teams tired of recreating analyses because nobody documented which version of which data went into last year's results.

Install to Claude Code

npx -y skills add davila7/claude-code-templates --skill lamindb --agent claude-code