Conciseness

The body is dense and expert-level (the "literal true" rule, 64 MiB cap, JSON vs structured-clone table, first-listener-wins) with no padding explaining basics Claude already knows. However, the `return true` and `toBeDefined()` points are each restated three times across How-to-use, Authoring, and the Anti-patterns table, and version-gated Chrome 148 detail adds time-sensitive material, so it is efficient with minor trimmable repetition rather than fully lean — the 4 anchor, not 5.

Actionability

Concrete, runnable sender/listener code for both one-shot shapes, real rejection-path assertions, a payload serialization table, and a worked regression-suite checklist; the three deeper shapes have executable recipes in the two verified reference files. Not 5: the flagship test example calls an undefined `sendFromContentScript` helper, so it is not copy-paste ready, and no test-runner/harness setup is shown — mostly executable with minor gaps, matching the 4 anchor.

Workflow Clarity

The 6-step How-to-use sequence is clearly ordered (identify shape → confirm running → write pair → assert full response → payload guardrails → re-check), with an explicit validation checkpoint ("Re-check the draft against the Anti-patterns table before committing") and a genuine feedback loop: each anti-pattern row pairs a failure with its fix, plus a full checklist-style worked example. This matches the 5 anchor's explicit validation, error-recovery fixes, and checklists; the 4 anchor would leave validation implicit or partial.

Progressive Disclosure

Clean split: the two simple one-shot shapes stay inline while the three heavier shapes are offloaded with well-signaled links to references/long-lived-ports.md and references/cross-boundary-messaging.md, both of which exist, contain substantive recipes, and nest no further local references (verified one level deep). The References section restates each link with content descriptions, matching the 5 anchor exactly.

Specificity

"Tests runtime.sendMessage, tabs.sendMessage, long-lived ports, externally_connectable and native messaging on a running extension" names five concrete, specific actions that comprehensively cover every Chrome extension messaging shape. It matches the 5 anchor (multiple specific concrete actions, comprehensive coverage) rather than 4, which expects minor gaps in coverage.

Completeness

Explicitly answers both: what ("Tests runtime.sendMessage, tabs.sendMessage, long-lived ports, externally_connectable and native messaging") and when ("Use when a Chrome extension's messages reach no listener, sendResponse never fires, a port disconnects mid-test, or a web page's access must be proven") with concrete trigger phrases. Clearly matches the 5 anchor; the 4 anchor requires a less explicit 'when', which is not the case here.

Trigger Term Quality

Includes natural symptom phrases users would actually say ("messages reach no listener", "sendResponse never fires", "a port disconnects mid-test") plus API names, but misses common variations like "onMessage listener", "message passing", or "content script not receiving". Good coverage with a few natural terms missing fits the 4 anchor; not 5 because synonym coverage is not comprehensive.

Distinctiveness Conflict Risk

"Chrome extension's messages" plus named APIs (externally_connectable, native messaging) carve out a clear niche with distinct triggers and minimal conflict risk against generic testing or extension-load skills. Not 4: there is no meaningful overlap risk with closely related skills given the messaging-specific trigger symptoms.