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

77%

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 ./backend/cli/skills/biology/anndata/SKILL.md

Quality

Content

72%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 well-structured, highly actionable skill body with excellent progressive disclosure into five reference files. The main weaknesses are repeated code patterns across sections and the absence of validation checkpoints in the batch/workflow sections.

Suggestions

Consolidate the backed-mode, sparse-matrix, and strings_to_categoricals examples so each pattern appears once, cross-referencing from the other sections instead of re-demonstrating.

Add an explicit validation step to the batch-integration and scRNA-seq workflows (e.g., assert shape/alignment after ad.concat, or verify adata.isbacked and obs/var integrity before writing) to establish a feedback loop.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and code-heavy with domain-specific detail (the X/obs/var/obsm structure) that earns its place, but backed mode, sparse-matrix conversion, and strings_to_categoricals are each demonstrated in 3-4 separate sections, which is tightening-eligible padding rather than minor over-explanation.	3 / 5
Actionability	Nearly every section gives copy-paste-ready, executable Python (read_h5ad with backed='r', write_h5ad with compression, ad.concat with join/label/keys, scverse integration, to_memory chunking), and the common cases (create, read, write, subset, concatenate, large-data) are all covered with real API calls.	5 / 5
Workflow Clarity	Multi-step workflows are clearly numbered (e.g., the scRNA-seq analysis 1-5 sequence and batch integration), but they lack explicit validation/verification checkpoints and there is no validate-fix-retry feedback loop; per the rubric cap, batch operations without validation cannot score above 3.	3 / 5
Progressive Disclosure	All five reference bundle files (data_structure, io_operations, concatenation, manipulation, best_practices) are real and linked one level deep with clearly signaled 'See: references/X.md' pointers plus topic bullets, giving a clean overview-to-detail split that is easy to navigate.	5 / 5
	Total	16 / 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-routed description that clearly states what it is and when to use it while disambiguating from sibling scverse skills. It could be sharper by naming concrete operations and including the 'AnnData' keyword users naturally type.

Suggestions

Add the keyword 'AnnData' and a synonym like 'scRNA-seq' so the description matches the term users most naturally type.

Replace generic handling verbs with 2-3 concrete operations the skill supports (e.g., 'create, read, write, concatenate, and subset annotated matrices').

Dimension	Reasoning	Score
Specificity	Names the domain ('Data structure for annotated matrices in single-cell analysis') and 1-2 actions ('working with .h5ad files', 'integrating with the scverse ecosystem'), but these are generic handling verbs rather than the concrete operations (create, read, write, concatenate, subset) the skill actually covers, so it stops at the 'domain + 1-2 actions, not comprehensive' anchor.	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'), and adds concrete routing triggers that disambiguate use, matching the 'clearly answers both what AND when with concrete trigger phrases' anchor.	5 / 5
Trigger Term Quality	Good coverage of natural terms users say — '.h5ad files', 'single-cell analysis', 'scverse ecosystem', 'scanpy', 'scvi-tools', 'cellxgene-census' — with the file extension included, but it omits the most common keyword a user would actually type ('AnnData'/'anndata') and synonyms like 'scRNA-seq', so it is not comprehensive enough for a 5.	4 / 5
Distinctiveness Conflict Risk	The clause 'This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census' carves a clear niche and explicitly routes away from the most likely conflicting skills, giving minimal conflict risk.	5 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/biology/anndata/SKILL.md
Commit: 3a6c3a9

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.