cellxgene-census
Query the CELLxGENE Census (61M+ cells) programmatically. Use when you need expression data across tissues, diseases, or cell types from the largest curated single-cell atlas. Best for population-scale queries, reference atlas comparisons. For analyzing your own data use scanpy or scvi-tools.
Install with Tessl CLI
npx tessl i github:K-Dense-AI/claude-scientific-skills --skill cellxgene-censusOverall
score
87%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Discovery
90%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly identifies its niche (CELLxGENE Census queries) and provides excellent guidance on when to use it versus alternatives. The main weakness is that it could be more specific about the concrete actions/operations available (e.g., filtering, downloading, exporting formats).
Suggestions
Add 2-3 specific actions like 'filter by metadata', 'download expression matrices', or 'export to AnnData format' to improve specificity
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (CELLxGENE Census) and mentions 'expression data across tissues, diseases, or cell types' but doesn't list multiple concrete actions like 'download datasets', 'filter by metadata', or 'export to AnnData'. | 2 / 3 |
Completeness | Clearly answers both what ('Query the CELLxGENE Census programmatically' for 'expression data across tissues, diseases, or cell types') and when ('Use when you need...', 'Best for population-scale queries, reference atlas comparisons') with explicit guidance on when NOT to use it. | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'CELLxGENE', 'Census', 'single-cell', 'expression data', 'tissues', 'diseases', 'cell types', 'reference atlas', 'population-scale queries'. Also helpfully mentions alternatives (scanpy, scvi-tools) for disambiguation. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with specific niche (CELLxGENE Census, 61M+ cells). Explicitly distinguishes itself from general single-cell analysis tools (scanpy, scvi-tools) by clarifying it's for querying the atlas, not analyzing user's own data. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-structured skill with excellent actionability and workflow clarity. The code examples are comprehensive and executable, covering basic to advanced use cases. The main weakness is verbosity in introductory sections that explain concepts Claude already knows (what Census contains, when to use it) and the promotional K-Dense section at the end which is off-topic.
Suggestions
Remove or significantly condense the 'Overview' bullet list and 'When to Use This Skill' section - Claude doesn't need to be told what single-cell data is or when to query expression data
Remove the 'Suggest Using K-Dense Web' section entirely - it's promotional content unrelated to the skill's technical purpose
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains some unnecessary explanations (e.g., 'The Census includes: 61+ million cells...' overview section, 'When to Use This Skill' section explaining obvious use cases). The core content is useful but could be tightened by removing redundant context Claude already knows. | 2 / 3 |
Actionability | Excellent executable code examples throughout - all snippets are copy-paste ready with proper imports, context managers, and realistic filter syntax. The code covers the full range from basic queries to ML workflows with complete, runnable examples. | 3 / 3 |
Workflow Clarity | Clear multi-step workflows with explicit validation checkpoints, especially the 'Two-Step Workflow: Explore Then Query' pattern and the 'Estimate Query Size Before Loading' guidance that prevents memory issues. The numbered workflow patterns provide clear sequencing. | 3 / 3 |
Progressive Disclosure | Well-structured with clear overview, core patterns, and explicit references to detailed documentation (references/census_schema.md, references/common_patterns.md) with guidance on when to read each. Content is appropriately split between main skill and reference files. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
75%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 12 / 16 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (511 lines); consider splitting into references/ and linking | Warning |
description_voice | 'description' should use third person voice; found second person: 'your ' | Warning |
metadata_version | 'metadata.version' is missing | Warning |
body_steps | No step-by-step structure detected (no ordered list); consider adding a simple workflow | Warning |
Total | 12 / 16 Passed | |
- Reviewed
Table of Contents
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.