unibind-database

Queries the UniBind database for experimentally validated transcription factor (TF) binding sites. Use when retrieving direct TF-DNA interaction datasets, downloading binding site coordinates (BED/FASTA) for local analysis, or listing available datasets by species, cell line, or TF name. Don't use to query specific intervals, locations, genes, motif models or expression data.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

86%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured, actionable, and free of condescending explanation, with executable commands and clear anti-patterns. The main weaknesses are minor redundancy between Core Rules and Quick Start and an inconsistency between the jq and jp guidance for large outputs.

Suggestions

Reconcile the large-output tooling: Quick Start says 'pipe through jq' while §3 says 'use uvx --from jmespath jp' — pick one and reference it consistently.

Trim the overlap between the Quick Start prose and the 'Use the Wrapper'/'Output' Core Rules so guidance is stated once.

Add a brief verification checkpoint for download_tfbs (e.g., confirm files written to --output-dir) to strengthen the workflow sequence.

Dimension	Reasoning	Score
Conciseness	Largely lean and assumes Claude's competence (no concept-explaining fluff), but the 'Use the Wrapper'/'Output' Core Rules overlap with Quick Start prose and the inconsistent jq (Quick Start) vs jp (§3) guidance adds mild clutter.	2.5 / 3
Actionability	Provides fully executable commands ('uv run <SKILL DIR>/scripts/unibind_api.py list_species'), concrete filter arguments with example values ('Homo sapiens', 'CTCF'), and copy-paste-ready invocations.	3 / 3
Workflow Clarity	Prerequisites are sequenced with conditional logic (license file absent -> notify -> create) and utility scripts are numbered, but there are no explicit validation/verification checkpoints; given the operations are read/download (low destructive risk) this sits between 2 and 3.	2.5 / 3
Progressive Disclosure	A self-contained overview organized into well-labeled sections with body references (scripts/unibind_api.py) resolving to real bundle files and single-level depth, fitting a moderately sized skill.	3 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 concise, concrete, and complete: it states what the skill does, when to use it with natural trigger terms, and where it stops with explicit exclusions. It is written in third person with no fluff or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('Queries... binding sites', 'retrieving... datasets', 'downloading binding site coordinates (BED/FASTA)', 'listing available datasets by species, cell line, or TF name'), matching the score-3 anchor for multiple specific concrete actions.	3 / 3
Completeness	Explicitly answers what (queries UniBind for validated TF binding sites) and when (a clear 'Use when...' clause with several triggers), matching the score-3 anchor with explicit triggers.	3 / 3
Trigger Term Quality	Covers natural terms a user would say — 'transcription factor', 'TF binding sites', 'BED/FASTA', 'species, cell line, or TF name' — giving good coverage rather than just jargon.	3 / 3
Distinctiveness Conflict Risk	A clear niche (direct TF-DNA interactions) plus a 'Don't use to query... motif models or expression data' exclusion makes it unlikely to trigger for the wrong skill.	3 / 3
	Total	12 / 12 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/unibind_database/SKILL.md
Commit: 0b42509

Reviewed: 8 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.