The Agent Skills Directory

[DYNAMIC_EXECUTION]: The PowerShell scripts scripts/check-vpn-lab-status.ps1 and scripts/watch-az-resource-state.ps1 act as wrappers that dynamically resolve and execute scripts located in a relative parent directory (../../../../scripts/) using Join-Path and the call operator (&).
[COMMAND_EXECUTION]: The skill is designed to execute Azure CLI (az) commands for infrastructure provisioning, configuration, and monitoring within the user's Azure environment.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted data which is subsequently used in CLI operations.
Ingestion points: Processes user-provided tenant ID, subscription ID, and target topology as defined in SKILL.md and references/validation-runbook.md.
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present in the prompt templates.
Capability inventory: Extensive resource management capabilities via Azure CLI (az) and local PowerShell execution.
Sanitization: No evidence of input sanitization or validation for the Azure identifiers or topology descriptions is provided in the instructions.

azure-infra-validation