anndata

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.

Quality

75%

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 ./skills/anndata/SKILL.md

Quality

Content

68%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.

A highly actionable, well-structured skill with comprehensive executable examples and properly signaled bundle references. The main weaknesses are duplicated inline command blocks across sections and missing validation checkpoints in the batch/destructive workflows.

Suggestions

Add explicit validation checkpoints to the batch-integration and large-dataset workflows (e.g., verify shapes/keys after ad.concat and assert file integrity after write_h5ad) so workflow_clarity can exceed the validation cap.

De-duplicate the per-section "Common commands" blocks against the Quick Start and the referenced reference files, keeping SKILL.md as an overview and leaving full command catalogs in the references.

Move or contextualize the version/release-date detail ("0.12.16, released 2026-05-18") into a compatibility/versioning note rather than the main Installation prose, to reduce time-sensitivity friction.

Dimension	Reasoning	Score
Conciseness	The prose is mostly lean and domain-specific, but read/write/subset/concat commands are duplicated across the Quick Start and the per-section "Common commands" blocks, and a time-sensitive release date ("0.12.16 (released 2026-05-18)") sits outside any deprecated/old-patterns section, so it could be tightened rather than fitting the efficient 4 anchor.	3 / 5
Actionability	The body is full of copy-paste-ready, executable code covering creation, reading, writing, subsetting, concatenation, integration, and full multi-step workflows, matching the fully-executable top anchor.	5 / 5
Workflow Clarity	Workflows (single-cell RNA-seq, batch integration, large datasets) are clearly numbered and sequenced, but batch/destructive operations like concatenation and file writing lack explicit validate-then-fix feedback loops, so the validation cap holds this at 3.	3 / 5
Progressive Disclosure	Structure is clear with one-level-deep, well-signaled references (all five references/*.md files exist) and per-section bullet lists of what each covers, but the inline "Common commands" blocks duplicate content that also lives in the referenced files, a minor organization gap keeping it just below the cleanly-split 5 anchor.	4 / 5
	Total	15 / 20 Passed

Description

82%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, well-targeted description that clearly states what the skill is, when to use it, and how it differs from sibling scverse skills. The main gap is specificity of concrete actions, which is implied rather than enumerated.

Dimension	Reasoning	Score
Specificity	Names the domain (single-cell analysis) and the core artifact (data structure for annotated matrices) plus a couple of concrete actions ("working with .h5ad files", "integrating with the scverse ecosystem"), but does not list multiple specific concrete actions like read/write/concatenate, so it falls short of the comprehensive-action anchors at 4-5.	3 / 5
Completeness	Explicitly answers both "what" (data structure for annotated matrices in single-cell analysis) and "when" (Use when working with .h5ad files or integrating with the scverse ecosystem) with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Includes strong natural trigger terms and the .h5ad file extension plus ecosystem tool names (scverse, scanpy, scvi-tools, cellxgene-census), but omits common synonyms such as zarr or single-cell RNA-seq, keeping it just below the comprehensive 5 anchor.	4 / 5
Distinctiveness Conflict Risk	The explicit disambiguation ("This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census") gives it a clear niche with minimal conflict risk, matching the top anchor.	5 / 5
	Total	17 / 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: K-Dense-AI/scientific-agent-skills
Path: skills/anndata/SKILL.md
Commit: 1e5eeff

Reviewed: 3 days 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.