CtrlK
BlogDocsLog inGet started
Tessl Logo

scanpy

Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, and visualization. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use anndata.

79

1.51x
Quality

75%

Does it follow best practices?

Impact

82%

1.51x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/scanpy/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

54%

14%

Single-Cell QC Pipeline for Tumor Microenvironment Study

QC filtering and metrics

Criteria
Without context
With context

MT- prefix identification

100%

100%

calculate_qc_metrics with qc_vars

100%

100%

filter_cells min_genes threshold

50%

100%

filter_genes min_cells threshold

100%

100%

MT% cell filter

0%

100%

QC violin plot

0%

0%

QC violin jitter

0%

0%

Normalize to 1e4

0%

0%

Log transform

0%

0%

Save raw counts

0%

0%

Save filtered output

100%

100%

Scatter QC plots

0%

0%

94%

26%

Downstream Analysis for PBMC Single-Cell Dataset

Normalization to clustering pipeline

Criteria
Without context
With context

HVG n_top_genes

100%

100%

Regress out covariates

0%

100%

Scale max_value

100%

100%

PCA arpack solver

100%

100%

Neighbors n_neighbors

0%

100%

Neighbors n_pcs

0%

100%

Leiden not Louvain

100%

100%

Multiple resolutions

100%

100%

Wilcoxon method

100%

100%

Save h5ad

100%

100%

Export metadata CSVs

100%

100%

PCA variance ratio plot

0%

0%

100%

46%

Publication Figure Suite for Single-Cell Manuscript

Publication-quality visualization

Criteria
Without context
With context

DPI 300 setting

100%

100%

PDF format

100%

100%

frameon=False

0%

100%

figures directory

100%

100%

use_raw=True for gene expression

0%

100%

legend_loc on data

0%

100%

Multiple marker genes per type

100%

100%

Dot plot included

100%

100%

Heatmap included

100%

100%

facecolor white

0%

100%

legend_fontoutline

0%

100%

Repository
K-Dense-AI/claude-scientific-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

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.