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

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 ./skills/human_protein_atlas_database/SKILL.md

Quality

Content

76%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.

The content is highly actionable with executable examples for every command and good progressive disclosure to a real bundle, but workflow sequencing and validation checkpoints are mostly implicit. Tightening the end-to-end flow and adding explicit validation would lift the weakest dimensions.

Suggestions

Add an explicit numbered end-to-end workflow (resolve Ensembl ID -> query by ID -> parse output) with a validation checkpoint confirming the result is non-empty and well-formed.

Trim redundancy between the 'When to Use' list and the 'Command Selection Guide' to improve token efficiency.

Include a short feedback loop for the search-hpa command (e.g., broaden/narrow the query and re-run) given its large, batch-like output.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with executable commands and minimal padding; the brief HPA-vs-RNA-seq context is useful rather than filler, though the 'When to Use' and 'Command Selection Guide' sections are partially redundant.	4 / 5
Actionability	All five subcommands are documented with concrete, copy-paste-ready 'uv run' invocations and complete argument lists, covering the common cases.	5 / 5
Workflow Clarity	Prerequisites and a troubleshooting section are present, but the core multi-step flow (resolve Ensembl ID then query by ID) is only implied, and there are no explicit validation checkpoints confirming output correctness before use.	3 / 5
Progressive Disclosure	The bundle is real and well-used (scripts/hpa_cli.py throughout, references/search-api.md clearly signaled at one level deep), with a well-organized overview-plus-command structure; minor organization gaps keep it just below a 5.	4 / 5
	Total	16 / 20 Passed

Description

66%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.

The description is clear, on-niche, and includes an explicit 'Use when' trigger, but it is somewhat generic in its single action and lacks a richer spread of natural trigger terms. It is solidly above average but not exemplary.

Suggestions

Add 1-2 more concrete actions (e.g., 'map gene symbols to Ensembl IDs', 'search genes by expression criteria') to raise specificity.

Expand trigger phrases with natural terms users say, such as 'protein levels', 'IHC staining', or 'where a protein is located in the cell'.

Offer a couple of distinct 'when' clauses (e.g., distinguishing protein-vs-RNA use) to make the trigger guidance more concrete.

Dimension	Reasoning	Score
Specificity	Names the concrete domain ('semi-quantitative protein expression and spatial localisation data from the Human Protein Atlas') and the action 'retrieve', but only one or two concrete actions are given, so coverage is not comprehensive.	3 / 5
Completeness	It states both what it does ('retrieve... data from the HPA') and when to use it ('Use when you want to retrieve...'), but the 'when' is a single generic clause rather than multiple concrete trigger phrases, keeping it just below a 5.	4 / 5
Trigger Term Quality	Relevant keywords like 'protein expression', 'spatial localisation', 'Human Protein Atlas', and 'HPA' appear, but common natural variations a user would say (e.g., 'protein levels', 'IHC', 'where is the protein located') are missing.	3 / 5
Distinctiveness Conflict Risk	It carves out a clear niche (HPA protein data) distinct from adjacent RNA-seq/GTEx skills, with triggers unlikely to fire for the wrong skill.	5 / 5
	Total	15 / 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: google-deepmind/science-skills
Path: skills/human_protein_atlas_database/SKILL.md
Commit: 6883275

Reviewed: 4 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.