Conciseness

Everything is task-specific with no generic concept padding, but there is genuine repetition that could be tightened: the 'not a fast overview / router tradeoff' point appears in the intro, again across the two-paragraph 'How it differs from a quick research router' section, and the core mechanisms (triangulation, parallel sub-agents, refresh protocol) are each restated in the phase table, 'Core mechanisms', and 'Anti-Patterns'.

Actionability

Concrete artifacts and thresholds throughout — plan.md, sources/NN_slug.md, refresh_targets.md, '≥3 independent, differently-typed sources per thesis', '2–4 falsifiable hypotheses', an exact output tree, and Credibility/Recency/Bias scoring. Not a 5 because key executable details are unspecified: the '4 self-critique questions' are never listed, no scoring scale is defined, and genre building blocks are only named.

Workflow Clarity

The 9-phase table gives a clear sequence with explicit validation checkpoints — Phase 4 're-evaluate between rounds', Phase 5 flagging claims as 'insufficient evidence', Phase 6.5 'Verify — lightweight citation check', Phases 5–6 confirming or refuting each hypothesis — and the anti-patterns act as an error-recovery checklist ('An empty fetch produces an empty claim, never a fabricated citation').

Progressive Disclosure

No bundle files exist and the ~85-line body is a well-organized self-contained overview with clearly signaled cross-references to sibling skills; however, detail such as the source-scoring rubric, the 4 self-critique questions, and genre block definitions is omitted rather than split into a one-level-deep reference file, leaving minor organization gaps.

Specificity

Lists multiple specific concrete actions — 'fan-out web search across many channels, parallel sub-agents, source triangulation (each claim backed by ≥3 independent sources), an adversarial review pass, and every source saved to its own file with verbatim quotes' — with comprehensive coverage of the workflow, in third person.

Completeness

Explicitly answers what ('Run a disciplined, multi-source research investigation... fan-out web search... adversarial review pass... every source saved to its own file') and when ('Use when a low-quality answer is expensive: strategy work, comparing N products/methods/markets, validating a hypothesis with external data, or mapping how a field works'), with concrete trigger phrases plus NOT-for boundaries.

Trigger Term Quality

Natural phrases like 'strategy work', 'comparing N products/methods/markets', 'validating a hypothesis', 'mapping how a field works', and 'quick fact-checks' match what users would actually say, but common synonyms such as 'deep dive', 'due diligence', or 'investigate' are missing, so coverage is good rather than comprehensive.

Distinctiveness Conflict Risk

Explicit disambiguation — 'NOT for quick fact-checks (answer directly), structured 12-dimension competitor scoring (use competitive-teardown), or fast topic overviews where the decision risk is low (use the research router instead)' — establishes a clear niche with minimal conflict risk.