CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

encode-ccres-database

Query the ENCODE Registry of cis-Regulatory Elements (cCREs) via the SCREEN GraphQL API, or make custom queries to the ENCODE Portal REST API for experiments and files (ChIP-seq peaks, etc.). Use when you want to query regulatory annotations or raw experimental data across human cell types.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%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 highly actionable, well-structured skill body: every command has executable examples and deep material is properly delegated to real reference files. The main weaknesses are a verbatim-duplicated parsing warning and the absence of any result-validation step after queries.

Suggestions

State the 'do not use cat / use jq' rule once — the 'Parsing Output (CRITICAL)' section repeats the Core Rules bullet almost word-for-word; keep the fuller version and trim the bullet to a pointer.

Fix the typo 'eachmcommand' in the Parsing Output section ('returned by eachmcommand').

Add a brief validation step after queries, e.g., check the output JSON is non-empty or inspect the error field before parsing, so failed lookups are caught early.

DimensionReasoningScore

Conciseness

Mostly efficient, but the 'Do NOT use cat / use jq' warning appears nearly verbatim twice (Core Rules bullet and the 'Parsing Output (CRITICAL)' section), and the intro re-explains domain background ('functional non-coding DNA elements (like Promoters, Enhancers, and insulators)') that could be trimmed. Not anchor 2 because there is no pervasive padding and the command reference itself is tight.

3 / 5

Actionability

Every command comes with a copy-paste-ready invocation with real arguments and output paths (e.g., 'uv run scripts/screen_api.py search --chromosome chr11 --start 5205263 --end 5207263 --output /tmp/search.json'), plus a worked jq filtering example. Not below 5 — the examples cover the common cases for both scripts.

5 / 5

Workflow Clarity

Prerequisites are sequenced first (uv setup, license notification), then usage rules, quick start, and per-command examples, with a documented fallback (jq unavailable → Python; grch38 → mm10). Not anchor 5 because there is no explicit validation step for query results (e.g., checking the output file is non-empty or handling failed lookups) — though as a read-only query skill this is a minor gap rather than a destructive-operation risk.

4 / 5

Progressive Disclosure

The body is a well-organized overview of commands, with bulk detail correctly split into one-level-deep, clearly signaled references ('read references/json_output_structure.md', 'read references/graphql_schema.md'), and both referenced files plus the scripts exist in the bundle. Not below 5 — the inline command docs are the appropriate getting-started content, not misplaced bulk material.

5 / 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: it names both API surfaces concretely, includes an explicit use-when clause, and occupies a distinct niche with minimal conflict risk. The main improvement opportunity is richer natural trigger phrasing (enhancers, promoters, ENCODE data) in the when-clause.

DimensionReasoningScore

Specificity

Names concrete actions — 'Query the ENCODE Registry of cis-Regulatory Elements (cCREs) via the SCREEN GraphQL API' and 'make custom queries to the ENCODE Portal REST API for experiments and files (ChIP-seq peaks, etc.)' — with specific data types. It stays below anchor 5 because only two API surfaces are named rather than a comprehensive list of specific operations (e.g., searching by coordinates, orthologs, linked genes).

4 / 5

Completeness

Clearly answers 'what' (query cCREs via SCREEN GraphQL; custom ENCODE Portal REST queries) and has an explicit 'Use when you want to query regulatory annotations or raw experimental data across human cell types' clause. The 'when' is present and explicit but slightly generic — it names object categories rather than the concrete user phrasings (e.g., 'ENCODE data', 'enhancers', 'ChIP-seq') that anchor 5 requires.

4 / 5

Trigger Term Quality

Good natural keyword coverage: 'ENCODE', 'cCREs', 'regulatory annotations', 'ChIP-seq peaks', 'experiments and files', 'human cell types' — terms a genomics user would naturally say. Not anchor 5 because common variations like 'enhancers', 'promoters', 'SCREEN', or 'genomic regions' are absent.

4 / 5

Distinctiveness Conflict Risk

A clear niche — the ENCODE cCRE registry via SCREEN GraphQL and the ENCODE Portal REST API — with distinct trigger terms unlikely to collide with other skills; no other common skill would trigger on these phrases.

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/encode_ccres_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.