Conciseness

The body is dense with routing signal and contains no explanations of concepts Claude already knows, but distinct-from material is repeated three times (frontmatter description, intro paragraph, and the dedicated 'Distinct from' section), and the 'When to invoke' symptom table overlaps the signal table. This sits between anchors 3 and 4 — the duplication (~15 of 145 lines) is slightly more than 'minor instances that could be trimmed' but well short of 'several unnecessary explanations'.

Actionability

Concrete guidance throughout: exact onboarding commands ('python3 skills/<sub-skill>/scripts/onboard.py --show'), a copy-paste forcing-question template ('Q1/1: ... Recommended: ... (Confirm, or override?)'), a deterministic signal table mapping keywords to sub-skills, a ≤200-word digest spec, and named artifact files. However, commands are templated on '<sub-skill>' and the referenced scripts, canon, and the 'Path-B' plan live outside this bundle and cannot be verified — concrete commands with minor gaps, i.e. anchor 4 rather than fully copy-paste-ready anchor 5.

Workflow Clarity

Five explicitly sequenced steps (explore → one forcing question → depth-first decision walk → forked invoke → digest) with confirmation checkpoints at every decision point: 'Confirm, or override?', 'Confirm before chaining', 'Never silently chain', 'Never run a sub-skill until the lane-defining decision is locked', plus an anti-patterns checklist. No destructive or batch operations exist, so the validation cap does not apply — this matches anchor 5's explicit validation steps and checklists.

Progressive Disclosure

Well-sectioned with clear headers, and per-lane detail (onboarding question sets, forcing-question libraries, canon) is correctly deferred to the four separate sub-skills with signaled paths ('skills/<sub-skill>/scripts/onboard.py'). Gaps: the References section name-drops canon (ICH E9, CONSORT, Bessemer) without tying it to actual files, and 'documentation/implementation/research-ops-expansion-plan.md' is an unresolvable external path; no bundle files exist in this skill to anchor navigation. Good structure with minor organization gaps — anchor 4.

Specificity

The description lists multiple concrete actions — 'planning, funding, scoping, or synthesizing enterprise research', 'Forks context to route to one of four Research-Operations sub-skills... and returns a digest' — with all four workstreams enumerated (clinical, R&D finance, market, product). Coverage is comprehensive with no gaps, matching the anchor-5 example rather than anchor 4; voice is consistently third person ('Forks', 'routes'), so no person penalty applies.

Completeness

'Use when planning, funding, scoping, or synthesizing enterprise research across workstreams' explicitly answers when with concrete trigger phrases, and 'Forks context to route to one of four... sub-skills... and returns a digest' explicitly answers what. Both questions are answered clearly and explicitly, matching anchor 5 exactly.

Trigger Term Quality

Twelve quoted natural phrases users would actually say — 'design this clinical study', 'what sample size', 'burn rate', 'capitalize or expense', 'TAM SAM SOM', 'market sizing', 'survey design', 'plan user interviews', 'usability test' — spanning all four lanes with synonym clusters (TAM/SAM/SOM + market sizing + segment the market). This matches the comprehensive-with-synonyms anchor 5; a few phrases like 'clinical trial' appear only in the body, which keeps it from being unquestionably exhaustive but still squarely at the top anchor.

Distinctiveness Conflict Risk

'Distinct from ra-qm-team (regulatory submission), finance (corporate close/valuation), research/grants (funding discovery), product-team (persona/journey/live experiments), and marketing-skill (campaign analytics)' affirmatively enumerates five neighbor boundaries. The niche (research-operations orchestration) is clear with distinct triggers and actively disambiguated overlaps — anchor 5 rather than 4, where overlap would merely be unaddressed.