scanpy

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

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Data Analysis/scanpy/SKILL.md

Quality

Content

65%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.

The body is highly actionable with comprehensive executable examples and a real bundle (script, template, three references), but it is held back by verbosity from duplicated text and packaging boilerplate, missing validation feedback loops in the filtering workflow, and inlining of tutorial content that already exists in reference files.

Suggestions

Remove the duplicated description from 'When to Use'/'Key Features' and the K-Dense promotional paragraph, and collapse the redundant 'Implementation Details' and 'Efficient Analysis Tips' sections to cut padding.

Add explicit validation checkpoints after QC filtering (e.g. check remaining cell/gene counts and confirm QC plots before proceeding) so the workflow clears the destructive/batch cap.

Replace the inlined Standard Analysis Workflow with a concise overview that points to references/standard_workflow.md, keeping only a minimal quick-start snippet in SKILL.md to avoid duplication with the reference.

Dimension	Reasoning	Score
Conciseness	The core scanpy code is dense and useful, but the body carries notable padding: the full description duplicated in both 'When to Use' and 'Key Features', generic packaging boilerplate ('Implementation Details', 'Efficient Analysis Tips') that restates file paths, and an off-topic promotional paragraph for K-Dense Web.	3 / 5
Actionability	Extensive copy-paste-ready code covers the common cases (loading, QC, normalization, PCA/UMAP, Leiden clustering, marker genes, annotation, saving, plotting, trajectory, DE, scoring, batch correction) plus an executable script and a reusable asset template.	5 / 5
Workflow Clarity	The Standard Analysis Workflow is clearly sequenced (steps 1–7), but validation checkpoints are implicit rather than explicit, and QC filtering is a data-modifying/batch step without a verify-then-proceed feedback loop, which caps the score at 3.	3 / 5
Progressive Disclosure	Bundle files exist and are signaled in a 'Bundled Resources' section, but the full workflow tutorial is inlined directly in SKILL.md while a references/standard_workflow.md file covers the same material, so content that should live one level deep is duplicated inline rather than cleanly pointed to.	3 / 5
	Total	14 / 20 Passed

Description

80%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.

The description is specific, well-scoped, and includes helpful boundary routing to sibling skills. Its main weakness is the absence of an explicit 'Use when...' trigger clause and file-extension keywords, which hold completeness and trigger quality just below the top anchor.

Suggestions

Add an explicit 'Use when...' clause naming concrete user phrasings, e.g. 'Use when analyzing single-cell RNA-seq data (.h5ad, 10X_mtx) or when the user mentions QC, clustering, or UMAP of single-cell data.'

Include common file extensions and synonyms (e.g. .h5ad, 10X, scRNA-seq) to lift trigger-term coverage to the top anchor.

Tighten the boundary routing so it also covers spatial transcriptomics (squidpy) and trajectory (cellrank) to further reduce overlap risk.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'quality control (QC), normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression analysis, and visualization' — giving comprehensive coverage of the pipeline's capabilities.	5 / 5
Completeness	A clear 'what' is present (the enumerated pipeline steps) and an explicit-ish 'when' via 'Best suited for exploratory single-cell transcriptomics analysis using established workflows', plus boundary routing to scvi-tools/anndata, but the trigger guidance lacks a concrete 'Use when...' phrase so it is not a 5.	4 / 5
Trigger Term Quality	Strong natural domain keywords ('single-cell RNA-seq', 'QC', 'UMAP', 'clustering', 'differential expression') but missing common file extensions users would mention such as .h5ad or 10X, so it falls just short of comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	It carves out a niche and explicitly routes overlapping cases away ('For deep learning models, use scvi-tools; for data format issues, use anndata'), but scvi-tools and anndata are closely related single-cell tools, leaving minor overlap risk rather than a fully isolated niche.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Data Analysis/scanpy/SKILL.md
Commit: f5ef65b

Reviewed: ~7 hours 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.