CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

human-protein-atlas-database

Use when you want to retrieve semi-quantitative protein expression and spatial localisation data from the Human Protein Atlas (HPA).

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

82%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 CLI skill: every command is executable as written, guidance on when not to use the skill is explicit, and detailed query syntax is properly deferred to a well-signaled reference file. Remaining gaps are minor: a redundant command-selection section, no explicit ordered resolve-to-query workflow, and one unreferenced bundle file.

DimensionReasoningScore

Conciseness

The body is efficient: each command section is a purpose line, a runnable example, and its arguments, with no padding or explanations of concepts Claude already knows. Not 5 because the Command Selection Guide restates the purpose of every Commands subsection, and the intro paragraph partially duplicates the description's RNA-vs-protein framing.

4 / 5

Actionability

Every subcommand has a copy-paste-ready `uv run scripts/hpa_cli.py ...` invocation with concrete arguments, the Quick Start covers the common resolve-then-query path, and Common Errors gives specific remediation steps (consult references/search-api.md, search the web for example queries, process output with jq). Fully executable with the common cases covered.

5 / 5

Workflow Clarity

The Ensembl-first dependency is explicit ("HPA endpoints are strictly Ensembl-based"), the Command Selection Guide maps user intent to the right subcommand, and Common Errors supplies recovery guidance for empty or oversized results. Not 5 because there is no explicit ordered resolve-to-query sequence or verification step confirming the symbol resolved before querying.

4 / 5

Progressive Disclosure

The body is a clean overview that appropriately pushes search query syntax into references/search-api.md, signaled clearly at both point of need (the search-hpa --query docs) and in Common Errors, exactly one level deep; scripts/hpa_cli.py is referenced throughout. Not 5 because references/citation.bib exists in the bundle but is never referenced or navigated from SKILL.md, a minor organization gap.

4 / 5

Total

17

/

20

Passed

Description

73%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 solid, domain-specific description with a clear Use-when clause and low conflict risk. Its main weaknesses are limited action coverage (only retrieval of two data types is mentioned) and missing common trigger synonyms such as the US spelling of localization and IHC-related terms.

Suggestions

Enumerate the skill's concrete actions more fully, e.g. "Map gene symbols to Ensembl IDs, retrieve tissue and cancer IHC expression levels, find subcellular locations, and search for genes matching expression criteria".

Add natural trigger synonyms such as "subcellular localization", "IHC/immunohistochemistry", and "Human Protein Atlas (HPA)" spelled the way users type it.

Replace the circular when-clause with concrete triggers, e.g. "Use when the user asks where a protein is expressed or located in the cell, or mentions HPA, protein expression, or IHC staining".

DimensionReasoningScore

Specificity

The description names the domain and 1-2 concrete actions ("retrieve semi-quantitative protein expression and spatial localisation data") but omits other capabilities the skill provides (gene-symbol-to-Ensembl mapping, searching genes by expression criteria, tissue and cancer-type breakdowns). It is not score 4 because it does not list several specific actions with only minor gaps.

3 / 5

Completeness

Both what ("retrieve semi-quantitative protein expression and spatial localisation data") and when ("Use when you want to...") are present. Not 5 because the when-clause is quasi-circular (use it when you want to retrieve what it retrieves) rather than concrete trigger phrases like "when the user mentions protein expression or HPA".

4 / 5

Trigger Term Quality

"Human Protein Atlas (HPA)", "protein expression", and "spatial localisation" are natural terms users would say, giving good keyword coverage. It falls short of 5 because common synonyms are missing: "subcellular localization" (US spelling), "IHC"/"immunohistochemistry", and "tissue expression".

4 / 5

Distinctiveness Conflict Risk

"Human Protein Atlas (HPA)" plus protein-level expression and localisation data carves out a clear niche with distinct triggers that would not fire for RNA-seq or general bioinformatics skills. Minimal conflict risk.

5 / 5

Total

16

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

google-deepmind/science-skills

Path

skills/human_protein_atlas_database/SKILL.md

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.