Name: tooluniverse-single-cell
Rating: 69.6 (1 reviews)
Author: mims-harvard

tooluniverse-single-cell

Single-cell RNA-seq analysis with scanpy/anndata — h5ad data loading, scRNA-seq quality control and QC gating (n_genes_by_counts, total_counts, mitochondrial percent / pct_counts_mt, pct_counts_ribo, doublet detection with Scrublet/scDblFinder, ambient RNA / SoupX awareness, empty-droplet filtering, MAD-based thresholds), normalization, dimensionality reduction (PCA, UMAP, t-SNE), clustering (Leiden, Louvain), marker gene identification, cell-type annotation, pseudotime/trajectory analysis. Use for any scRNA-seq workflow, including deciding which cells to filter, flag, or investigate before downstream analysis.

Quality

83%

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

The canonical home for this skill is tooluniverse-single-cell in mims-harvard/ToolUniverse

Quality

Content

75%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 executable code, concrete commands, and a sequenced QC-first workflow with validation guards. The main weaknesses are over-explanation of biology Claude already knows and a dangling analysis_patterns.md reference whose file is absent from the bundle.

Suggestions

Trim the biology rationale under 'WHY each metric flags a problem' and the Reasoning Framework to the non-obvious decisions only; Claude already knows why mitochondrial leakage indicates stress.

Create references/analysis_patterns.md (referenced three times in the decision tree and Reference Documentation) or repoint those links to an existing file so all referenced paths resolve.

Collapse the Scanpy vs Seurat equivalents table into the seurat_workflow.md reference, since it is lookup material rather than workflow guidance Claude needs inline.

Dimension	Reasoning	Score
Conciseness	Mostly efficient code-and-table content, but several sections explain biology Claude already knows (e.g., why high mito implies a dying cell via membrane rupture, ribo being cell-type-specific) and the Reasoning Framework and Scanpy-vs-Seurat table restate well-known concepts, fitting the score-3 anchor of mostly efficient with some unnecessary explanation.	3 / 5
Actionability	Copy-paste-ready code blocks for loading, QC metrics, the full pipeline, DE, statistics, and Harmony, plus concrete CLI invocations (python scripts/scrna_qc.py --install-plan, the tu run run_deseq2_analysis command) covering the common cases as required by the score-5 anchor.	5 / 5
Workflow Clarity	A clear decision tree sequences QC -> Normalize -> HVG -> PCA -> Cluster -> Annotate -> DE with explicit validation checkpoints (Rule Zero pre-computed check, 'HONEST EXECUTION' install-plan guard, 'Never report cutoffs you did not actually run'), but a few checkpoints are implicit rather than enumerated validate->fix->retry loops, so it sits at 4 rather than 5.	4 / 5
Progressive Disclosure	Good one-level-deep structure: the body is an overview pointing to eight real references/ files (scanpy_workflow, scrna_qc, etc.) and scripts/ with clear signaling, but it references analysis_patterns.md which does not exist in references/, and one reference path is inconsistent, leaving minor organization gaps at the score-4 level.	4 / 5
	Total	16 / 20 Passed

Description

92%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 highly specific, well-scoped description that names many concrete scRNA-seq actions and pairs them with explicit 'Use for' trigger guidance and clear boundaries against sibling skills. The only minor gap is synonym coverage in the trigger phrasing.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions (h5ad loading, QC gating with named metrics like n_genes_by_counts and pct_counts_mt, doublet detection, normalization, PCA/UMAP/t-SNE, Leiden/Louvain clustering, marker identification, annotation, pseudotime) — comprehensive domain coverage matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers both what (the enumerated analysis capabilities) and when ('Use for any scRNA-seq workflow, including deciding which cells to filter, flag, or investigate before downstream analysis'), matching the score-5 anchor that requires concrete trigger phrases for both.	5 / 5
Trigger Term Quality	Strong natural terms (scRNA-seq, h5ad, QC, UMAP, marker genes, cell-type annotation) and file formats (h5ad, 10X), but a few common user phrasings like 'single cell', 'clustering', or 'which cells to filter' synonyms are partially covered rather than exhaustively enumerated, placing it just below the score-5 synonym-saturation anchor.	4 / 5
Distinctiveness Conflict Risk	A clear single-cell niche with distinct triggers and the body explicitly routes overlapping cases (bulk RNA-seq DESeq2, enrichment, VCF) to other skills, giving minimal conflict risk as required by the score-5 anchor.	5 / 5
	Total	19 / 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: mims-harvard/ToolUniverse
Path: plugins/tooluniverse/skills/tooluniverse-single-cell/SKILL.md
Commit: 1b02920

Reviewed: 1 day 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.