Conciseness

Mostly efficient domain-specific gotchas (ploidy defaults, 2-row VarSeq headers, coding denominators) that earn their tokens, but at ~500 lines the body is noticeably padded: the coding-only denominator rule is stated three times (PRIMARY SCRIPTS table, CRITICAL section, Analysis Conventions), multi-row Excel headers are explained twice, and the CRISPR section re-explains PAM/NGG basics Claude already knows. Fits 'mostly efficient but includes some unnecessary explanation or could be tightened'; not 2 because the bulk is genuinely non-obvious operational knowledge rather than generic filler.

Actionability

The three primary scripts get exact copy-paste bash invocations with flags, workdir, and named output keys to grep ('SNP_COUNT_ALLELES', 'AVERAGE_PER_SAMPLE'), which is fully executable. However, the Python pattern examples call functions (answer_vaf_mutation_fraction, variant_analysis_pipeline, parse_vcf, FilterCriteria) that are not defined in any bundled script and whose provenance (ToolUniverse tools vs local code) is never made explicit — a minor gap. Matches 'Mostly executable guidance; concrete code or commands with minor gaps'; not 5 because those examples cannot be run as written.

Workflow Clarity

Sequencing is clear and prioritized (RULE ZERO check-first → PRIMARY SCRIPTS → domain recipes → phase workflow) with some real checkpoints (the ~0.4 synonymous-fraction sanity check with a recompute loop, 'PLOIDY=<value>' emitted for confirmation, report-all-three-conventions sensitivity). Fits 'Clear sequence with most checkpoints present; minor validation gaps'; not 5 because there are three competing entry paths (phases, scripts-first, RULE ZERO) with no single explicit order among them, and the phase workflow itself lacks validation steps.

Progressive Disclosure

Good structure: four real one-level-deep reference files (references/vcf_filtering.md, mutation_classification_guide.md, annotation_guide.md, sv_cnv_analysis.md — all verified present) clearly signaled inline and in a Reference Documentation section, plus six bundled scripts documented with a usage table. Minor gaps keep it from 5: 'QUICK_START.md' is referenced under Additional Resources but does not exist in the bundle, and a fair amount of reference-grade material (full tool tables, phase-by-phase detail) is inlined in an already-long SKILL.md.

Specificity

Quotes like "parsing, annotation, classification (synonymous, missense, frameshift, stop_gained), VAF filtering, coding vs non-coding categorization, multi-condition variant comparison" list multiple concrete, comprehensive actions with named variant types — matching the anchor 'Lists multiple specific concrete actions; comprehensive coverage'. Not 4 because coverage has no meaningful gaps; the parenthetical SO terms make the actions unusually concrete.

Completeness

The 'what' is explicit and detailed ("VCF and variant analysis — parsing, annotation, classification..., VAF filtering...") and the 'when' is an explicit 'Use for' clause with concrete triggers ("Use for VCF parsing, variant fraction calculations..., and per-sample mutation profiling"). This matches the anchor 'Clearly and explicitly answers both what AND when with concrete trigger phrases'; the explicit trigger guidance also avoids the ≤3 cap.

Trigger Term Quality

"VCF parsing", "variant fraction calculations", "per-sample mutation profiling", and the named consequence terms are natural phrasings, but common synonyms/extensions users would say are missing — e.g. ".vcf", "SNV/indel", "variant calling". Matches 'Good keyword coverage; a few natural terms missing'; not 5 because the synonym/extension coverage is not comprehensive, and not 3 since far more than 'some relevant keywords' are present.

Distinctiveness Conflict Risk

"VCF", "VAF filtering", "coding vs non-coding categorization", and named SO consequence terms carve out a clear niche (variant analysis) with distinct triggers and minimal overlap with adjacent genomics skills. Not 4 because no closely related skill's triggers plausibly collide with these terms.