Lean and operational — assumes Claude knows the services, delegates heavy detail to references, and avoids teaching basics; the minor restatement of 'paper compliance' is deliberate reinforcement of a critical anti-pattern rather than padding.

Gives concrete directives (scope IAM roles to needed actions, enforce TLS via aws:SecureTransport, bound blast radius with a stop condition) but defers all executable CLI commands to references and leaves some troubleshooting guidance directional ('Expand blast radius, add multi-fault scenarios').

The Define→Test→Operate sequence is explicit, and the 'validate findings before you resolve them' section provides a hard validation checkpoint ('Run the experiment, confirm the system recovers... then mark resolved') for a consequential operation, with troubleshooting recovery paths.

Overview body points to three real, one-level-deep references (lifecycle-workflow.md, best-practices.md, api-reference.md) with clearly signaled links and contextual labels; heavy API and workflow content is appropriately split out.

Lists multiple concrete actions across the lifecycle: 'policy creation through failure mode assessment, to FIS experiment validation, to ARC operational controls' — specific, not vague.

Explicitly answers both what (integrated Define→Test→Operate lifecycle across three services) and when ('Applicable when...', 'Also applicable for...', 'Not applicable for...'), with explicit triggers in third person.

Uses natural domain terms a user would say ('complete resilience strategy', 'planning a resilience program', 'validate findings before resolving them', 'are they done') alongside the real service vocabulary (NGRH, FIS, ARC).

Clear three-service niche with explicit boundary against the sibling skill 'resilience-hub-failure-mode-assessment' and single-service queries, making wrong-skill triggering unlikely.