Dense, token-efficient formatting using tables and terse bullets with no padding explaining concepts Claude already knows; every section earns its place rather than the 'mostly efficient but could be tightened' score-2 anchor.

Provides concrete, copy-paste-ready guidance — naming templates (META_Conv_Lookalike-Customers_FreeTrial_2024Q1), second-by-second video structure, 70/30 budget splits, and numbered optimization levers — far exceeding the 'some concrete guidance but incomplete' score-2 anchor.

Clear sequenced flow from context-gathering through platform selection, structure, a pre-launch checklist with validation checkpoints, to optimization and reporting, with conditional feedback loops ('If CPA is too high → 1. Check landing page...') and explicit checklists.

The body is an overview that points to three real, one-level-deep reference files (ad-copy-templates.md, audience-targeting.md, platform-setup-checklists.md) with clear 'See [references/...]' signaling, verified to exist on disk.

Lists multiple concrete capabilities — 'campaign strategy, ad creation, audience targeting, and optimization' — alongside specific platforms, matching the 'lists multiple specific concrete actions' anchor rather than the partial score-2 anchor.

Explicitly answers both what ('This skill covers campaign strategy, ad creation, audience targeting, and optimization') and when ('Also use when the user mentions...'), with an explicit 'Use when' clause present.

Covers natural terms users actually say — 'PPC,' 'paid media,' 'ad copy,' 'ad creative,' 'ROAS,' 'CPA,' 'retargeting,' 'audience targeting' — plus platform names, giving good coverage rather than just 'some relevant keywords.'

A clear paid-advertising niche with named platforms and distinct triggers (PPC, ROAS, paid media) makes it unlikely to fire for the wrong skill, matching the 'clear niche with distinct triggers' anchor.