cellxgene-census

Programmatically query the CZ CELLxGENE Census (61M+ cells) when you need cross-tissue, disease, or cell-type expression data for population-scale queries and reference atlas comparisons.

Quality

83%

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

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 complete runnable example and clear sequencing, though somewhat redundant in its Implementation Details. The main weakness is progressive disclosure: the existing reference bundle files are never linked from the skill body.

Suggestions

Link the existing bundle files from the body, e.g. under an Advanced section: 'Schema details: See [census_schema.md](references/census_schema.md)' and 'Common query patterns: See [common_patterns.md](references/common_patterns.md)'.

Trim Implementation Details sections that re-explain what the commented example already demonstrates (filtering semantics, expression-layer columns) to improve conciseness.

Add a lightweight validation checkpoint for the out-of-core path, such as confirming the query returns rows before iterating or handling an empty result.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with a complete runnable example and tight section headers, but the Implementation Details section re-explains filtering semantics, access-pattern choice, and expression-layer columns already shown in the commented code, which could be trimmed.	4 / 5
Actionability	Provides a complete, copy-paste-ready Python script covering metadata exploration, in-memory AnnData loading, and out-of-core iteration, plus exact install commands — fully executable across the common cases.	5 / 5
Workflow Clarity	The example lays out a clear numbered 1-4 sequence (open pinned census, explore metadata, load AnnData, out-of-core query) with version pinning for reproducibility, but lacks explicit validation/error-recovery checkpoints for the large-scale iteration path; the destructive/batch cap does not apply since this is read-only querying.	4 / 5
Progressive Disclosure	The body has good section structure, but the two provided bundle files (references/census_schema.md, references/common_patterns.md) are never referenced or linked from SKILL.md, leaving them orphaned, and portions of Implementation Details could live in those references rather than inline.	3 / 5
	Total	16 / 20 Passed

Description

87%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 clearly states what the skill does and when to use it with concrete trigger phrases and a distinct niche. Minor gaps are the lack of common single-cell synonyms and the fact that all capabilities fold into one query action.

Dimension	Reasoning	Score
Specificity	Names the domain ("CZ CELLxGENE Census (61M+ cells)") and the concrete action ("Programmatically query") plus several specific use modalities (cross-tissue, disease, cell-type expression, population-scale queries, reference atlas comparisons), but all reduce to one core query action rather than a list of distinct operations.	4 / 5
Completeness	Explicitly answers both "what" ("Programmatically query the CZ CELLxGENE Census") and "when" ("when you need cross-tissue, disease, or cell-type expression data for population-scale queries and reference atlas comparisons") with concrete trigger phrases, satisfying the Use-when requirement.	5 / 5
Trigger Term Quality	Strong natural terms for the target audience ("cross-tissue", "disease", "cell-type expression data", "reference atlas", "population-scale queries"), but omits common synonyms users would also say such as "single-cell", "scRNA-seq", or "scRNA".	4 / 5
Distinctiveness Conflict Risk	The "CZ CELLxGENE Census (61M+ cells)" niche is highly specific with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	18 / 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/Evidence Insight/cellxgene-census/SKILL.md
Commit: 63c61d3

Reviewed: 10 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.