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

71%

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

61%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 body with strong executable examples and clean one-level-deep references that all resolve to real files. It loses points on duplicated inline content across sections and on multi-step workflows (concatenation, filtering, normalization) that include no validation or verification checkpoints.

Suggestions

Add validation/verification steps to the batch-oriented workflows (e.g., print adata.shape and obs value counts after concat, confirm adata.isbacked and X dtype after backed reads, reopen and sanity-check written h5ad files) to lift workflow clarity above the batch-operation cap.

Deduplicate content that appears in multiple sections — h5ad read/write (Quick Start vs Core Capabilities), concat examples (section 3 vs Batch integration workflow), and sparse/backed-mode advice (section 5 vs Troubleshooting) should appear once with the rest deferred to the reference files.

Fix or remove non-executable snippets: the AnnCollection example should pass AnnData objects rather than filename strings, and the `process(chunk)` placeholders in the large-datasets workflow should be replaced with real operations.

Dimension	Reasoning	Score
Conciseness	Mostly efficient but with clear duplication: h5ad read/write appears in both Quick Start and Core Capabilities, concatenation examples in section 3 and the Batch integration workflow, and sparse/backed-mode advice in both section 5 and Troubleshooting. The Overview also explains package history Claude already knows. Not 2 (no tutorial-style padding); not 4 given the real repetition across sections.	3 / 5
Actionability	Code examples are concrete and largely executable (ad.read_h5ad with backed mode, ad.concat with join/label/keys arguments, subsetting, full scanpy pipeline). Kept below 5 by the AnnCollection example passing filename strings where AnnData objects are expected, and by the `process(adata_subset)` placeholder calls in the large-datasets workflow.	4 / 5
Workflow Clarity	Workflows are numbered and clearly sequenced (e.g., the 5-step RNA-seq workflow), but batch and destructive operations (batch concatenation, QC filtering, normalization) lack validation checkpoints — nothing verifies results after concat, filtering, or writing. The rubric caps workflow clarity at 3 for batch operations without validation steps.	3 / 5
Progressive Disclosure	All five referenced files exist in references/, are one level deep, and are clearly signaled with "See: `references/...`" plus per-file content bullets. Scored against the actual bundle structure this is good, but the inline Integration, Common Workflows, and Troubleshooting sections carry reference-level detail that could be split out, keeping it below the top anchor.	4 / 5
	Total	14 / 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 description with an explicit when-clause, concrete file-extension triggers, and excellent sibling-skill disambiguation. Its main weakness is that the what-clause names the artifact type without listing any concrete actions the skill performs (read, write, concatenate, subset).

Dimension	Reasoning	Score
Specificity	The description names the domain and artifact type ("Data structure for annotated matrices in single-cell analysis") but lists no concrete actions such as creating, reading, or concatenating AnnData objects. It is more concrete than a generic domain mention (2) but does not list the several specific actions required for 4.	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's structure. The when-clause is specific, so this is not the level below where the when is only weakly implied.	5 / 5
Trigger Term Quality	Includes natural triggers like ".h5ad files", "single-cell", "scverse ecosystem", plus ecosystem tool names (scanpy, scvi-tools, cellxgene-census). Missing a few natural variations such as "anndata" itself, "scRNA-seq", or "single-cell RNA-seq", keeping it short of comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	The description actively disambiguates against sibling skills ("This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census"), establishing a clear niche with 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: 4082a2e

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.