Conciseness

The body is mostly operational (bash commands, search queries, tables, a brief template), but includes several passages that could be trimmed: 'The insight: ... steal their design decisions', 'Why This Matters ... Most developers only use 20% of their platform', 'The difference between "build a note app" and ... is knowing what the platform offers', and enumerated lists of well-known libraries (TanStack, React Hook Form, Zod) that Claude already knows. This lands on the mostly-efficient-but-could-be-tightened anchor rather than the minor-trimmage anchor at 4.

Actionability

Guidance is directly executable throughout: copy-paste bash for local exploration ('find ~/Documents -maxdepth 2 -name "CLAUDE.md" ...'), concrete search query patterns per research area ('best X', 'X vs Y', 'top [product] plugins'), research-question tables with named sources (bundlephobia, changelog URLs), and a complete fill-in research-brief template. This matches the fully-executable anchor at 5.

Workflow Clarity

A clear 10-step sequence (Understand Intent → Local Exploration → Web Research → Ecosystem → Competitor → Library → Platform → Future-Casting → Technical → Synthesis) with depth-gated steps labeled '(wide + deep)' and a default depth stated. However, there are no explicit checkpoints or feedback loops (e.g. confirm depth/intent with the user, or criteria for when research is sufficient), which keeps it at 4 rather than the validation-with-feedback-loops anchor at 5.

Progressive Disclosure

The 300+ line body has good section headers but is monolithic — no references/, scripts/, or assets/ bundle exists, and substantial self-contained content (the ~60-line Cloudflare capability checklist, the ~60-line brief template, the competitor deep-dive table) clearly belongs in separate one-level-deep reference files. Structure is present but content that should be split is inlined, matching the anchor at 3 rather than 4 where 'most content is appropriately placed'.

Specificity

Lists multiple specific concrete actions — 'Explores local projects for reusable code, researches competitors, reads forums and reviews, analyses plugin ecosystems, investigates technical options, and produces a comprehensive research brief' — with three named depth levels and their durations. Coverage spans the full workflow from local scan to synthesis with no clear gaps, so it matches the comprehensive anchor rather than the minor-gaps anchor at 4.

Completeness

Explicitly answers both what (the enumerated research actions and the deliverable research brief) and when ('Triggers: ... before I start building', 'this is pre-build product discovery'), matching the top anchor exactly. It even adds a negative boundary, which exceeds the anchor rather than falling to 4.

Trigger Term Quality

The explicit 'Triggers:' clause gives seven natural phrases users would actually say ('research this', 'discovery', 'what should I build', 'competitive analysis', 'before I start building', 'research before coding'). Good coverage, but a few natural synonyms are missing (e.g. 'market research', 'look into', 'scoping'), which keeps it at the good-coverage anchor rather than the comprehensive-with-synonyms anchor at 5.

Distinctiveness Conflict Risk

Clear niche (pre-build product discovery) with distinct triggers and an explicit disambiguation: 'Not for cited fact-checking research reports (a separate harness does those)'. This minimizes conflict risk with general research skills, matching the clear-niche anchor at 5.