The body is lean and dominated by executable TypeScript that earns its tokens; it does not explain what a service worker is and keeps prose to actionable notes, with only a brief rhetorical opener that does not pad meaningfully.

Every section ships copy-paste-ready, complete TypeScript (persistent-context setup, offline worked example, evaluate, cache-first, network-first) with real Playwright APIs rather than pseudocode.

The "How to use" section sequences the procedure, and the worked example numbers its steps with validation checkpoints (waitForEvent before asserting, waitForLoadState('networkidle')); operations are not destructive/batch so no explicit fix-retry loop is required.

A concise overview points to one clearly signaled, one-level-deep bundle file — references/advanced-service-worker-tests.md (verified present) — for advanced recipes, keeping the main skill appropriately split.

Lists multiple concrete actions — "Test service workers with Playwright (`context.serviceWorkers()` + `waitForEvent('serviceworker')`)", "unit tests via `service-worker-mock`", cache strategies, and "`evaluate()` continuity across worker restart" — matching the multiple-specific-actions anchor.

Explicitly answers both what (Playwright + service-worker-mock testing, lifecycle, cache strategies, evaluate continuity) and when ("Use when a site registers a service worker and its caching / offline behavior is uncovered, or users report stale content surviving a deploy").

Uses natural terms a user would say — "registers a service worker", "caching / offline behavior", "stale content surviving a deploy" — giving good coverage rather than only technical jargon.

Clear niche with explicit disambiguation — routing install/add-to-homescreen to pwa-install-flow-tests and cache-policy design to sw-cache-strategy-author — making conflict with sibling skills unlikely.