Conciseness

Efficient, table-driven, and assumes Claude's competence — no beginner explanations of Azure or IaC concepts. Minor trimming possible: the mermaid flowchart largely duplicates the prose pipeline description in the same section ('Every phase advances only after its gate passes' is stated twice). Not 5 due to that duplication; not 3 because there is no padding or unnecessary explanation beyond it.

Actionability

Concrete and executable: names specific MCP tools with purposes, real CLI commands ('az deployment group create', 'az bicep build', 'terraform init/plan/validate/apply', 'checkov'), exact artifact paths ('<project-root>/.azure/infrastructure-plan.json', 'infra/main.bicep + infra/modules/*'), and specific fixes in the Error Handling table. Not 5 because the body stops at naming commands rather than showing a worked example (delegated to references); not 3 because every tool, command, path, and fix is specific.

Workflow Clarity

The 7-phase pipeline is clearly sequenced with explicit validation checkpoints and feedback loops: Phase 5 requires explicit user approval, Phase 6 requires 'pass local validation (az bicep build / terraform validate) with zero errors, pass a checkov security scan', and the flowchart shows error loops (VAL -- errors --> P6) plus a risk-acknowledged deploy confirmation at Phase 7. Destructive deployment is gated by explicit confirmation. Matches the top anchor: explicit validation steps, error-recovery loops, and a checklist-style completion self-check.

Progressive Disclosure

SKILL.md is a lean overview; all detail is delegated to real, verified reference files with clear navigation (Quick Reference table, 'Workflow (Start Here)', resources/ and constraints/ README indexes). Not 5 because references run two levels deep (SKILL.md → workflow.md → phases/*.md, and resources/README.md → category files), which is deeper than the one-level-deep ideal — though each intermediate file is itself a clear index table, keeping navigation easy. Not 3 because nothing that belongs in a reference is inlined and every reference is clearly signaled.

Specificity

Lists several concrete actions — 'Architect and provision enterprise Azure infrastructure', 'Generates Bicep or Terraform directly (no azd)' — and names concrete domains ('networking, identity, security, compliance, and multi-resource topologies with WAF alignment'). Not 5 because actions stay at the domain level (e.g., it doesn't enumerate concrete deliverables like plans, module files, or validation output); not 3 because multiple specific actions are explicitly stated.

Completeness

Explicitly answers both questions: 'what' via 'Architect and provision enterprise Azure infrastructure from workload descriptions... Generates Bicep or Terraform directly', and 'when' via a dedicated WHEN: clause with concrete trigger phrases. This matches the top anchor exactly; a score of 4 would require the 'when' to be less explicit, which it is not.

Trigger Term Quality

The WHEN clause supplies seven natural user phrasings with verb variety — 'plan Azure infrastructure', 'architect Azure landing zone', 'design hub-spoke network', 'plan multi-region DR topology', 'set up VNets firewalls and private endpoints'. Not 5 because a few natural terms are missing (e.g., plain 'Terraform' requests, 'hub and spoke' spelled out, general cloud-migration phrasings); not 3 because coverage spans several distinct entry points rather than a single keyword.

Distinctiveness Conflict Risk

Clear niche (enterprise infrastructure planning/provisioning) with distinct triggers, plus explicit disambiguation — 'PREFER azure-prepare FOR app-centric workflows' and '(no azd)'. Minimal conflict risk; matches the top anchor for a clear niche with distinct triggers.