Conciseness

The body is dense and lean — tables for query strategies, fallback chains, error responses, and quality tiers, with no padding or explanation of concepts Claude already knows. It is not a 5 because 'Prioritize >=3 biological replicates, complete annotations, and both raw+processed data' and the RNA-seq/microarray dynamic-range comparison appear verbatim in both 'Domain Reasoning' and 'Reasoning Framework', a trimmable duplication.

Actionability

Concrete parameterized tool calls (tu.tools.arrayexpress_search_experiments(keywords=..., species=..., limit=20)), fallback chains, and per-error response tables make the guidance mostly executable. It falls short of 5 because the main code block uses placeholders ('[gene/disease]', '[species]') rather than a worked example with real values — only the Bgee/GxA blurbs include concrete IDs and parameters.

Workflow Clarity

Phases 0-3 are explicitly sequenced with skip/ask conditions in Phase 0, disambiguation strategy in Phase 1, fallback chains in Phase 2, and a structured report with quality scoring in Phase 3. Error handling provides explicit feedback loops ('Broaden keywords, remove species filter, try synonyms'; 'Retry once'), and the Synthesis Questions act as a validation checklist, matching the score-5 anchor. Retrieval is read-only, so the destructive-operation cap does not apply.

Progressive Disclosure

The single file is well-organized with clear section headers and no nested references, and no bundle files exist. It is not a 5 because ~20 lines of extended tool-reference material (the long GxA, ARCHS4, and Bgee paragraphs plus Search Parameters) is inlined API documentation that could live in a references/ file, leaving the overview less lean than the anchor requires.

Specificity

The description lists multiple concrete actions covering the full workflow: 'Retrieve gene expression and omics datasets', 'gene disambiguation and quality assessment', 'finding RNA-seq/microarray datasets by organism/tissue/condition', 'comparing across studies', and 'assessing dataset suitability before downloading'. Coverage is comprehensive rather than leaving minor gaps, so it fits the score-5 anchor rather than 4.

Completeness

It explicitly answers both questions: the 'what' ('Retrieve gene expression and omics datasets from ArrayExpress and BioStudies with gene disambiguation and quality assessment') and an explicit 'Use for...' trigger clause with concrete scenarios. This matches the score-5 anchor exactly; score 4 would require the 'when' to be less explicit, which it is not.

Trigger Term Quality

Strong natural keyword coverage including 'gene expression', 'RNA-seq', 'microarray', 'case-control, time-series, dose-response', and named databases (ArrayExpress, BioStudies). A few natural terms users would say are missing, such as 'transcriptomic', 'differential expression', or 'GEO', so it falls just below the comprehensive-synonyms anchor at 5 but well above 3.

Distinctiveness Conflict Risk

A clear niche (omics dataset retrieval) with distinct triggers naming specific databases (ArrayExpress, BioStudies) and technologies (RNA-seq/microarray). Minimal conflict risk with generic data-analysis skills; nothing pushes it below the score-5 anchor.