About eight lines with no concept padding — only the executable command, the validation behavior, and the auth-error recovery; every token earns its place and assumes Claude's competence.

Provides a copy-paste-ready executable invocation with parameter syntax ('.\scripts\runAzCommand.ps1 -Command "<az command>"'); not below 3 because it is concrete and complete rather than pseudocode.

For a simple single-task skill the sequence is unambiguous: run the script, it validates against allowed prefixes and exits with an error on failure, then handle auth errors via 'az login' — an explicit validation checkpoint and recovery loop are present.

A short single-purpose skill with a single one-level-deep bundle reference (scripts/runAzCommand.ps1, verified to exist) that is clearly signaled; not below 3 because the split is appropriate and there is no nested referencing.

Lists multiple concrete actions — 'checking authentication, querying Log Analytics, managing deployments, or inspecting Azure resources' — matching the top anchor for naming several specific capabilities.

Explicitly answers what ('Runs Azure CLI commands in a terminal') and when ('Use this skill when you need to execute az CLI commands such as...'), matching the both-what-and-when anchor; not below 3 because the 'Use when' trigger is explicit, not implied.

Natural terms users would say are well covered: 'Azure CLI commands', 'az CLI', 'authentication', 'Log Analytics', 'deployments', 'Azure resources'; not below because coverage is broad rather than missing variations.

A clear niche (Azure CLI execution) with distinct triggers unlikely to overlap with unrelated skills; not below 3 because the domain is specific rather than generic.