Conciseness

The body is two lines — a generator comment and 'Load and follow the shared skill at `~/.agents/skills/plan-design-review/SKILL.md`' — with zero padding and no explanation of concepts Claude already knows. Every token earns its place, matching the 5 anchor ('lean and efficient; assumes Claude's competence').

Actionability

The single instruction is concrete and executable (an exact file path to load), but the body contains no task-level guidance — none of the seven passes, the 0-10 rating scale, or the editing steps the description promises. This fits anchor 3 ('some concrete guidance but incomplete; missing key details') rather than 4, where the gaps would be minor.

Workflow Clarity

The multi-step process the description advertises ('seven passes', rating, editing) is entirely absent from the body — only a single delegating 'Load and follow' directive is present, with no sequence and no validation for the actual work. It sits between anchor 1 (steps missing) and anchor 2 (rough sequence, many gaps); the lone directive is coherent and well-defined, so 2 rather than 1.

Progressive Disclosure

The body is a clearly signaled, one-level-deep pointer, but it targets an absolute path outside the skill bundle (`~/.agents/skills/plan-design-review/SKILL.md`) and no references/, scripts/, or assets/ files exist, so the body provides no overview and the skill ships none of its content. Fits anchor 3 ('some structure but could be better organized') rather than 4, since the organization gap is not minor.

Specificity

It enumerates multiple concrete actions — 'seven passes over information architecture, interaction states, emotional arc, AI-slop risk, design-system alignment, responsive and accessibility coverage, and unresolved decisions', each 'rated 0-10', then 'the plan is edited to get there' — comprehensive coverage matching the 5 anchor. Not 4, because there are no meaningful gaps in the enumerated actions.

Completeness

Both questions are answered explicitly: the 'what' is the seven-pass rating-and-edit loop, and the 'when' is concrete — 'Use when a plan has a UI surface and you want the design gaps closed before code is written' — plus a negative scope ('Not for auditing a live site'). Not 4, because the when-clause is already explicit and trigger-phrase-level specific.

Trigger Term Quality

Natural terms like 'plan', 'design doc', 'UI surface', 'design gaps', and 'before code is written' are present, but common variations users would say — 'wireframe', 'mockup', 'spec', 'design review' — are missing. Matches anchor 4 (good keyword coverage, a few natural terms missing) rather than 5's comprehensive synonym coverage.

Distinctiveness Conflict Risk

The pre-implementation plan-review niche and the explicit 'Not for auditing a live site' boundary distinguish it well, but broad terms like 'review' and 'design' leave minor overlap risk with general design-critique or code-review skills. Fits anchor 4 (mostly distinct, minor overlap risk) rather than 5's minimal-conflict clear niche.