CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ensembl-database

Query the Ensembl database to resolve gene, transcript, and protein IDs, fetch genomic or protein sequences, retrieve gene structures (exons), and get variant consequence and effect predictions (VEP). Use this skill as a primary ID translator, genomic sequence database and variant effect prediction tool.

Quality

83%

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.

The body is a well-structured, highly actionable command reference: every operation has executable examples, defaults and user-facing flags are explicitly governed by Core Rules, and detailed endpoint material is correctly offloaded to a clearly signaled one-level-deep reference file. Remaining weaknesses are minor — a few trimmable concept definitions, no explicit error-recovery guidance, and one unmentioned bundle file.

DimensionReasoningScore

Conciseness

The body is dominated by commands, flags, and runnable examples that earn their tokens, with only minor over-explanation: the 'Key Concepts' section defines ENSG/ENST/ENSP stable-ID basics Claude already knows, and the Overview paragraph partially repeats the frontmatter description. This matches the 4 anchor (efficient with minor trimmable instances) better than 3, since there is no padded fluff or unsolicited commentary.

4 / 5

Actionability

Every command (resolve-gene, map-id, get-sequence, gene-summary, transcripts, canonical-tss, transcript-structure, protein-info, protein-sequence, vep) has copy-paste-ready 'uv run scripts/ensembl_api.py ...' invocations with realistic arguments and useful variants (symbol vs. ENSG input, --assembly GRCh37, --only-mane), plus a concrete VEP stdout sample and jq/python parsing guidance. This fully matches the 5 anchor: executable and covering the common cases.

5 / 5

Workflow Clarity

Commands chain naturally through consistent example IDs (TP53 → ENSG00000141510 → ENST00000269305) and the Core Rules impose explicit output checkpoints ('MUST explicitly flag this default', 'show the complete list'), giving a clear sequence. It falls short of 5 because there is no explicit multi-step recipe or error-recovery loop for API failures (e.g., what to do on a failed resolve or a 429), and it is above 3 since no destructive or batch operations are involved and checkpoints are present.

4 / 5

Progressive Disclosure

The full API endpoint reference is appropriately split into references/ensembl_rest_api_reference.md (verified to exist, 325 lines), clearly signaled in the Custom Queries section and one level deep, keeping the body a lean command overview. The minor gap keeping it at 4 rather than 5 is that references/citation.bib exists in the bundle but is never referenced in the body, leaving a small navigation hole.

4 / 5

Total

17

/

20

Passed

Description

78%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 that names four concrete capabilities and includes an explicit use-role clause, with good natural trigger terms and a clearly distinct genomics niche. Its main weaknesses are minor: missing external-database mapping terms (UniProt/RefSeq cross-referencing), absent synonym variations, and a 'when' clause that could use concrete user-side trigger phrases.

DimensionReasoningScore

Specificity

Lists four concrete actions ('resolve gene, transcript, and protein IDs', 'fetch genomic or protein sequences', 'retrieve gene structures (exons)', 'get variant consequence and effect predictions (VEP)'), which is strong but stops short of comprehensive — external-database cross-mapping (UniProt/RefSeq), gene summaries, and TSS lookup are not named. It exceeds the 3 anchor (only 1-2 actions) but has minor coverage gaps, matching the 4 anchor.

4 / 5

Completeness

The 'what' is explicit (four concrete capabilities) and 'Use this skill as a primary ID translator, genomic sequence database and variant effect prediction tool' provides explicit use-role guidance, so it is not capped at 3. However, it lacks concrete user-side trigger phrases (e.g., 'when the user mentions an ENSG ID or asks for variant consequences'), matching the 4 anchor where 'when' could be more explicit.

4 / 5

Trigger Term Quality

Good natural domain keywords users would say — 'Ensembl', 'VEP', 'gene IDs', 'variant consequence', 'genomic sequence', 'ID translator' — but common variations like 'cross-reference IDs', 'ID mapping', or 'lookup gene' are missing. Coverage is solid without being comprehensive of synonyms, fitting the 4 anchor.

4 / 5

Distinctiveness Conflict Risk

'Ensembl database', 'VEP', and gene/transcript/protein ID resolution carve a clear, domain-specific niche with distinct triggers and minimal conflict risk against non-genomics skills. It clearly matches the 5 anchor rather than the 4 anchor, whose overlap with closely related skills does not materialize here.

5 / 5

Total

17

/

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/ensembl_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.