The body is lean and avoids explaining concepts Claude already knows, but it restates overlapping territory three times — the 'What is covered' bullets, the 'When to use' list, and the Workflow steps cover similar ground — so it could be tightened further.

It gives concrete executable commands ('./mvnw compile', 'mvn clean verify') and specific file/class guidance, but the actual @ChangeUnit code and configuration examples are deferred to the reference rather than appearing inline, leaving the body's guidance incomplete on its own.

The four-step workflow is clearly sequenced with explicit validation checkpoints (MANDATORY compile-before, SAFETY stop-on-failure, VERIFY after) forming a real feedback loop, which is especially appropriate for destructive database migration operations.

The body is a concise overview pointing to a single, real, well-organized reference file one level deep (verified present with headed sections and examples), with the link clearly signaled — easy navigation and appropriate content split.

The description enumerates many concrete actions ('Mongock runner/driver selection, Micronaut bean wiring, migration scan packages, @ChangeUnit classes, lock/transaction settings, and Testcontainers verification') which would warrant a 3, but it is written in second person ('Use when you need to'), so the voice penalty reduces it by one to 2.

It explicitly answers both what it does (the enumerated capabilities) and when to use it via an explicit 'Use when...' clause plus concrete trigger examples, satisfying both halves.

It lists natural user phrasings ('Add Mongock migrations in Micronaut; Review Micronaut MongoDB data migrations; Configure Mongock change units with Micronaut Data MongoDB') that a user would plausibly say, giving good coverage of trigger variations.

The narrow niche of Mongock MongoDB migrations within Micronaut, with highly specific trigger terms, makes it unlikely to fire for unrelated skills.