scanpy

Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, visualization, and converting R-friendly single-cell formats such as Seurat or SingleCellExperiment RDS files into h5ad for Scanpy. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use anndata.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable content with a composable script toolkit, clear workflow sequencing, and clean progressive disclosure to real reference files. The main weakness is conciseness: the Quick Start and ancillary sections re-explain basics Claude already knows and overlap with the script table.

Suggestions

Trim the 'Understanding AnnData Structure' block (adata.X/obs/var/uns/obsm comments) — Claude already knows AnnData; link to the anndata skill instead.

Consolidate 'Key Parameters to Adjust' and 'Common Pitfalls' into the script table or a single reference, since they duplicate guidance already implied by the toolkit flags.

Add an explicit validate-then-proceed checkpoint in the step-by-step chain (e.g. run inspect_data.py after each stage and only proceed on expected shape/counts) to lift workflow clarity.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, but the Quick Start re-explains basic AnnData structure (adata.X, adata.obs) that Claude already knows, and the Key Parameters / Common Pitfalls sections partly duplicate guidance already covered by the script toolkit table.	3 / 5
Actionability	Fully executable: copy-paste-ready CLI invocations with concrete flags, a full script toolkit table with typical calls, and one-shot and step-by-step chains covering the common cases.	5 / 5
Workflow Clarity	The seven-step sequence, one-shot pipeline, and step-by-step chain give a clear path with inspect-based checkpoints, but the body lacks explicit 'validate then proceed' feedback loops for destructive/batch operations beyond relying on per-script --help.	4 / 5
Progressive Disclosure	SKILL.md is a well-signaled overview with one-level-deep references to real files (references/analysis_workflow.md, api_reference.md, plotting_guide.md, r_interop.md) and a clearly organized scripts/ and assets/ bundle, all of which exist.	5 / 5
	Total	17 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that names concrete capabilities, surfaces natural trigger terms including file extensions, and explicitly scopes when to use it versus neighboring skills. It answers both 'what' and 'when' clearly and avoids vague fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, visualization' — plus format conversion, giving comprehensive coverage of the scanpy capability surface.	5 / 5
Completeness	Explicitly answers 'what' (the pipeline steps) and 'when' via an explicit 'Use for ...' clause, with additional 'Best for ...' scoping and skill-routing guidance to scvi-tools and anndata.	5 / 5
Trigger Term Quality	Covers natural terms and synonyms a user would say — 'single-cell RNA-seq', 'QC', 'UMAP', 't-SNE', 'clustering', 'marker genes' — plus file extensions and format names (.h5ad, .rds, Seurat, SingleCellExperiment).	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Scanpy scRNA-seq) and explicitly disambiguates from scvi-tools (deep learning) and anndata (format questions), minimizing wrong-skill triggers.	5 / 5
	Total	20 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/scanpy/SKILL.md
Commit: 49c6e97

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.