anndata

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.

1.09x

Quality

70%

Does it follow best practices?

Impact

81%

1.09x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/anndata/SKILL.md

Evaluation results

84%

18%

Multi-Lab scRNA-seq Batch Integration

Multi-batch concatenation with source tracking

Criteria

Without context

With context

Concatenation with label

100%

Concatenation with keys

100%

Index uniqueness

100%

Join strategy selection

100%

Strings to categoricals

Raw data preservation

100%

Gzip compression

100%

Sparse matrix usage

100%

Processing metadata stored

Embedding name convention

100%

Correct concat axis

100%

82%

Scalable Single-Cell QC Pipeline for Atlas-Scale Data

Backed mode and memory-efficient processing

Criteria

Without context

With context

Backed read-only open

100%

Metadata inspection before load

100%

Filtered to_memory()

100%

Sparse matrix handling

100%

Backed read-write for metadata

80%

Processing history in uns

100%

Gzip compression on write

Strings to categoricals

No full dense conversion

100%

Chunked processing or subset

100%

78%

Genomics Data Ingestion and Format Standardization Pipeline

Format conversion and I/O optimization

Criteria

Without context

With context

MTX transpose applied

100%

External metadata index alignment

100%

Strings to categoricals before write

50%

100%

H5AD with gzip compression

100%

Zarr with chunks parameter

100%

Compression tradeoff documented

37%

25%

Sparse matrix for sparse data

100%

Var names uniqueness

Read elem usage

Processing log written

100%

Correct import alias

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Multi-Lab scRNA-seq Batch Integration Scalable Single-Cell QC Pipeline for Atlas-Scale Data Genomics Data Ingestion and Format Standardization Pipeline

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.