Tessl • The management layer for AI agent skills

SkillPublic

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

90%

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

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.

A well-structured, highly actionable command-menu skill with copy-paste-ready invocations and clear out-of-scope guidance. Its main weaknesses are a batch download path lacking a verification checkpoint (capping workflow clarity) and some duplicated output-handling guidance plus an unreferenced bundled citation file.

Suggestions

Add a verification step after `download_tfbs` (e.g., check expected file count and non-empty BED/FASTA, report failures) so the batch download has an explicit validation checkpoint.

De-duplicate the large-output / pipe-through-tool guidance: state it once (Core Rules or Quick Start) and reference it from section 3 instead of repeating it, and reconcile the `jq` (Quick Start/Anti-Patterns) vs `jp` (section 3) inconsistency.

Either reference references/citation.bib from the body (e.g., in a Citation section) or remove it from the bundle so the bundle structure matches what the skill surfaces.

Dimension	Reasoning	Score
Conciseness	Mostly lean, command-driven content, but the large-output / pipe-through-tool guidance is repeated in Quick Start, Core Rules, and section 3, and 'Output: Query commands print JSON' appears twice — efficient with minor duplication that could be trimmed, landing above the level-3 'could be tightened' anchor.	4 / 5
Actionability	Every section gives copy-paste-ready `uv run <SKILL DIR>/scripts/unibind_api.py ...` commands with concrete filter flags and a worked download example, covering the common cases per the level-5 anchor; the `<SKILL DIR>` placeholder is explicitly documented.	5 / 5
Workflow Clarity	Commands are organized by task and prerequisites are sequenced, but `download_tfbs` is a batch operation with no verification checkpoint (e.g., confirm file count/integrity), so per the batch-operations rule workflow_clarity is capped at 3 even though the menu is well laid out.	3 / 5
Progressive Disclosure	Good section structure (Prerequisites, Quick Start, Core Rules, Utility Scripts, Anti-Patterns) with one-level-deep references to scripts/unibind_api.py; minor gaps — references/citation.bib is bundled but never referenced in the body, and the inlined filter-flag list could live in a reference — keep it at level 4 rather than 5.	4 / 5
	Total	16 / 20 Passed

Description

100%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, third-person description that crisply states what the skill does, when to use it, and when not to, with concrete actions and natural trigger terms including file formats. No vague fluff or over-claims; the explicit out-of-scope boundaries sharply reduce conflict with sibling database skills.

Dimension	Reasoning	Score
Specificity	Names the domain and multiple concrete actions — 'retrieving direct TF-DNA interaction datasets, downloading binding site coordinates (BED/FASTA)... listing available datasets by species, cell line, or TF name' — giving comprehensive coverage rather than the 1-2 actions at level 3.	5 / 5
Completeness	Explicitly answers both 'what' ('Queries the UniBind database for experimentally validated TF binding sites') and 'when' ('Use when retrieving... downloading... or listing...') with concrete trigger phrases, matching the level-5 anchor rather than the level-4 'when could be more specific'.	5 / 5
Trigger Term Quality	Natural domain terms a genomics user would say ('transcription factor binding sites', 'TF-DNA interaction datasets', 'datasets by species, cell line, or TF name') plus file extensions ('BED/FASTA'), matching the comprehensive synonym-and-extension coverage anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (UniBind TF binding sites) with explicit 'Don't use to... motif models or expression data' boundaries steering users to adjacent skills, giving minimal conflict risk per the level-5 anchor.	5 / 5
	Total	20 / 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/unibind_database/SKILL.md
Revision: 4de6786

Reviewed: 1 month ago

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.

unibind-database

Table of Contents

Is this your skill?