Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

LaminDB Biological Data Lakehouse Assistant

A skill that guides Claude through LaminDB artifact registration, querying, lineage tracking, and ontology-backed validation for biological data.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

LaminDB is an open-source, lineage-native lakehouse for biology. This skill packages the knowledge Claude needs across six capability areas:

  • Artifact management: register versioned DataFrames, AnnData, Zarr, Parquet and more
  • Lineage tracking: ln.track() / ln.finish() and @ln.flow() / @ln.step() capture notebooks, scripts and pipeline runs
  • Querying: filter operators, cross-registry traversal, ln.Q logical queries, chunked streaming of large files
  • Curation & validation: DataFrameCurator, AnnDataCurator, SpatialDataCurator, TiledbsomaExperimentCurator with flexible-to-strict schemas
  • Bio ontologies: Bionty-backed genes (Ensembl), cell types (CL), tissues (Uberon), diseases (Mondo), pathways (GO)
  • Integrations & deployment: Nextflow, Snakemake, Redun, W&B, MLflow, Hugging Face, S3/GCS/MinIO, SQLite vs PostgreSQL

It also encodes safety defaults: never echo credentials, prefer IAM/env vars, sanitize external data before saving, and pin package versions.

Who it's for

  • Researchers handling scRNA-seq, spatial transcriptomics or multi-omics data
  • Bio data engineers who must prove reproducibility and provenance
  • Teams running Nextflow/Snakemake pipelines that need organized outputs
  • Groups standardizing metadata for FAIR compliance

Example uses

  1. Curate an scRNA-seq dataset: "Validate the cell_type column in this h5ad against the CL ontology and register it as an artifact."
  2. Build a queryable lakehouse: register dozens of batches with feature annotations, then retrieve all tissue="PBMC", condition="treated" datasets in one filter call.
  3. Link ML runs: store a trained model as an artifact tagged with the W&B run ID so the model traces back to its exact training data.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/lamindb folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/lamindb/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git /tmp/scientific-agent-skills && mkdir -p ~/.claude/skills && cp -r /tmp/scientific-agent-skills/skills/lamindb ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the skills repository: 
    git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Create the Claude skills directory and copy just the lamindb skill: 
    mkdir -p ~/.claude/skills
cp -r scientific-agent-skills/skills/lamindb ~/.claude/skills/
  4. Confirm the reference docs came along: 
    ls ~/.claude/skills/lamindb/references
  5. Restart Claude Code, then try a prompt like "Register this h5ad file in LaminDB with ontology validation."
  6. Before real use, install LaminDB in your Python environment and log in (check PyPI for the current stable version, then pin it): 
    uv pip install lamindb bionty
lamin login
lamin init --storage ./mydata
View source on GitHub ↗License: MIT