research-router-skill

Route broad or ambiguous life-sciences research requests to the right skills, normalize core entities, optionally parallelize independent evidence gathering with subagents when available, and synthesize a concise evidence-backed answer. Use when a user asks a general life-sciences question that could span multiple sources or analysis types.

Quality

88%

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

Quality

Content

92%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.

A highly actionable, concise orchestration skill with a clearly sequenced workflow and an explicit entity-stability gate. The only gap is progressive disclosure: everything lives inline in one file with no split-out reference material.

Suggestions

Move the long skill-selection heuristics and entity-normalization skill catalogs into a bundled reference file (e.g. references/skill-catalog.md) and link to it from SKILL.md, keeping the core routing logic lean.

Add a short 'Quick start' or minimal-routing section at the top so the core decision flow is visible before the detailed lane and skill tables.

Dimension	Reasoning	Score
Conciseness	Lean and well-sectioned with no padding or explanation of concepts Claude already knows; every list (lanes, entity-skill mappings, heuristics) carries concrete, decision-relevant content. Not penalized because nothing reads as filler.	3 / 3
Actionability	Gives concrete, executable guidance for an instruction-only router: named downstream skills per entity type and task type, explicit subagent delegation criteria, and a structured output contract. Specific enough to act on directly.	3 / 3
Workflow Clarity	A clear numbered sequence (understand → normalize → select → gather → synthesize) with an explicit gating checkpoint ('Do not start broad evidence collection until the important entities are stable enough to route correctly') and bounded subagent objectives. Not below 3 because the gate functions as a validation checkpoint even though this is not a destructive operation requiring fix-retry loops.	3 / 3
Progressive Disclosure	Well-organized into clear sections with no nested references, but it is a single monolithic file with no external references — the skill-selection catalogs and entity-mapping tables are inline material that could be split out. Not the level above because there are no one-level-deep well-signaled references, and not level 1 because organization and navigation are good and there is no deep reference nesting.	2 / 3
	Total	11 / 12 Passed

Description

85%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.

A strong, third-person description that specifies concrete actions and an explicit Use-when clause with a clear niche. Its main weakness is trigger-term coverage, which stays abstract and omits the natural user phrasings that would broaden recall.

Suggestions

Add natural user phrasings to the Use-when clause, e.g. 'Use when a user asks broad questions like "what is known about X" or mixes entities (gene plus disease, variant plus phenotype).'

Include a couple of concrete entity-pair trigger examples so the description mirrors the real-world request language rather than only 'multiple sources or analysis types'.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Route broad or ambiguous life-sciences research requests to the right skills, normalize core entities, optionally parallelize independent evidence gathering with subagents... and synthesize a concise evidence-backed answer' — matching the multi-action anchor.	3 / 3
Completeness	Clearly answers both what (route, normalize, parallelize, synthesize) and when ('Use when a user asks a general life-sciences question that could span multiple sources or analysis types') with an explicit trigger clause.	3 / 3
Trigger Term Quality	Has relevant trigger terms ('general life-sciences question', 'multiple sources', 'analysis types') but misses common natural variations a user would say, such as 'what is known about...', which appears only in the body, not the description. Not the level above because coverage of natural user phrasings is incomplete.	2 / 3
Distinctiveness Conflict Risk	Scoped to 'broad or ambiguous' and 'general life-sciences question that could span multiple sources', carving a clear orchestration niche distinct from individual downstream skills and unlikely to mis-trigger for narrow lookups.	3 / 3
	Total	11 / 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: openai/plugins
Path: plugins/life-science-research/skills/research-router-skill/SKILL.md
Commit: 11c74d6

Reviewed: 17 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.