The Agent Skills Directory

[Data Ingestion Surface]: The skill is designed to process external information from the advisor_remediation_get tool and public Microsoft Learn documentation to generate remediation steps. This is a common and intended pattern for documentation-based tools. To ensure security, the skill includes explicit instructions to attribute sources and avoid generating commands for general guidance.
[Command Generation Guidelines]: The instructions direct the agent to produce Azure CLI and PowerShell commands based on retrieved official sources. As a security best practice, the skill explicitly mandates that no commands should be executed without the user's explicit confirmation, maintaining human-in-the-loop control.
[Placeholder Handling]: The skill includes logic to substitute resource-specific values (like resourceId and subscription) into command templates. It specifies that the agent must use exact values and should never guess missing information, which helps prevent accidental operations on unintended resources.

azure-advisor-actionability