What problem does it solve? Engineering teams running Azure workloads often lack a systematic way to evaluate whether their code and infrastructure can survive zone failures and regional outages. This Skill runs a structured five-phase research-to-assessment workflow that inspects source code and IaC, producing evidence-based resiliency findings with file and line references and a prioritized P0-P3 remediation plan. ## Core Features & Use Cases - Five-Phase Workflow: Runs core research (Prompts 0-7), service-specific research for Azure services like AKS, Cosmos DB, SQL, Redis, and Key Vault (Prompts 8-19), consolidation, planning, and a final code-level resiliency assessment report. - Evidence-Only Research: Keeps research phases forensic with file and line citations, no remediation or code examples, and classifies every finding as P0, P1, P2, or P3. - Flexible Execution Modes: Supports Mode A (interactive, one prompt per turn with /clear between steps) and Mode B (autonomous subagent orchestration with up to three parallel service prompts). - Use Case: A team preparing an Azure multi-region active/active deployment invokes the skill to audit zone survivability and regional failover readiness, receiving a consolidated research document, a master remediation plan, and a code-level assessment report. ## Quick Start Run /hve-resiliency-research in Copilot Chat, choose Mode A or Mode B when prompted, and let the workflow begin at Phase 1 Prompt 0.
