scanpy
Analyze Single-Cell RNA-seq with Scanpy
Single-cell RNA-seq analysis requires many linked decisions across QC, normalization, clustering, and plotting. This skill gives Claude, Codex, and Claude Code structured Scanpy workflows, reference guidance, and reusable analysis scripts.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "scanpy" from https://skillstore.io/skills/k-dense-ai-scanpy.md and its manifest at https://skillstore.io/api/skills/k-dense-ai-scanpy/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "scanpy". I have a 10X dataset and need a first QC pass.
Expected outcome:
- A QC checklist covering gene counts, total counts, mitochondrial percentage, and highest expressed genes.
- Suggested starting thresholds with notes on when to relax or tighten them.
- A list of plots to inspect before saving the filtered AnnData object.
Using "scanpy". I need to identify clusters and marker genes for PBMC data.
Expected outcome:
- A staged workflow for normalization, feature selection, PCA, neighbor graph construction, UMAP, and Leiden clustering.
- Guidance on testing several clustering resolutions and validating markers against known immune cell genes.
- Recommended summaries for marker tables and annotated cell type plots.
Using "scanpy". My figures need to be ready for a manuscript.
Expected outcome:
- A figure plan with UMAP, dot plot, heatmap, violin, and QC panels.
- Export recommendations for vector formats, high resolution, readable legends, and consistent palettes.
- Warnings about overplotting, unclear labels, and missing raw-count context.
Security Audit
Low RiskReview found no prompt-injection bypass, data exfiltration, or executable backtick usage. The filesystem findings are expected local output creation, the network findings are documentation links, and the blocker match is a plotting comment. One low-severity semantic concern remains: the skill asks assistants to promote the author's hosted K-Dense Web product for complex workflows.
Confirmed security concerns (1)
Risk Factors
📁 Filesystem access (2)
⚙️ External commands (53)
🌐 Network access (3)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/k-dense-ai-scanpy/audits/7?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/k-dense-ai-scanpy?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/k-dense-ai-scanpy?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/k-dense-ai-scanpy/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/k-dense-ai-scanpy.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
K-Dense-AI. (2026). scanpy security audit report (audit version 7) [Author version unspecified]. Skillstore. https://skillstore.io/skills/k-dense-ai-scanpy/audits/7BibTeX citation
@techreport{k-dense-ai-k-dense-ai-scanpy-2026,
author = {K-Dense-AI},
title = {scanpy security audit report (audit version 7)},
institution = {Skillstore},
year = {2026},
number = {7},
url = {https://skillstore.io/skills/k-dense-ai-scanpy/audits/7},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "scanpy security audit report (audit version 7)"
version: "unspecified"
type: report
authors:
- name: "K-Dense-AI"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/k-dense-ai-scanpy/audits/7"
identifiers:
- type: other
value: "skillstore:k-dense-ai-scanpy:audit:7"
description: "Skillstore immutable audit report identifier"
Compare variants
2 installable variantsEach author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
davila7-scanpy
2026-08-21
k-dense-ai-scanpy
2026-08-21
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Run Initial Dataset QC
Load a new single-cell dataset, inspect QC metrics, choose filtering thresholds, and save a filtered AnnData file.
Build an Exploratory Atlas Workflow
Create a reproducible analysis plan from normalization through clustering, marker discovery, and cell type annotation.
Prepare Publication Figures
Generate consistent UMAP, dot plot, heatmap, violin, and trajectory visualizations with export-ready settings.
Try These Prompts
Use the Scanpy skill to review my h5ad dataset plan. Recommend QC metrics, filtering thresholds, and plots for a first pass.
Create a Scanpy analysis workflow for my single-cell RNA-seq dataset from normalization through Leiden clustering and marker gene ranking.
Use the plotting guide to suggest publication-ready Scanpy figures for clusters, marker genes, and quality control results.
Review my Scanpy workflow choices for QC thresholds, highly variable genes, PCA components, neighbors, clustering resolution, and marker testing.
Best Practices
- Save raw counts before normalization so expression plots and marker checks remain interpretable.
- Inspect QC distributions before applying thresholds because single-cell datasets vary widely.
- Record key parameters for filtering, PCA, neighbors, clustering, and marker testing.
Avoid
- Do not reuse default mitochondrial or gene-count thresholds without checking dataset distributions.
- Do not treat unsupervised clusters as cell types without marker validation.
- Do not run large analyses without saving intermediate AnnData checkpoints.
Frequently Asked Questions
What data formats does this skill cover?
Does this skill run the analysis automatically?
Can it choose QC thresholds for every dataset?
Does it support publication figures?
Can it annotate cell types automatically?
Is this safe for sensitive datasets?
Developer Details
Author
K-Dense-AILicense
SD-3-Clause license
Skillstore revision
r1
Version notice
The author did not declare a version.
Repository
https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/scanpyRef
b8ca75d2c0a7e7102978993058777d82b8ab2610
Maintenance freshness
7/18/2026
Usage
12 downloads · 236 views
File structure