The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Azure CLI (az) and the az rest command to interact with Azure infrastructure. These commands are used to query billing data, retrieve advisor recommendations, and search for orphaned resources via Resource Graph queries, which is appropriate for the skill's stated purpose.
[EXTERNAL_DOWNLOADS]: The skill references legitimate Azure-related extensions and tools, such as the resource-graph extension and the azqr (Azure Quick Review) tool from the official Azure GitHub organization. These are standard administrative tools for the target environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes JSON data returned from Azure APIs, creating a potential attack surface for indirect prompt injection if resource metadata were to contain malicious instructions.
Ingestion points: Data enters the agent's context through JSON responses from az rest, az advisor, and az graph commands.
Boundary markers: The skill relies on standard tool-output handling rather than defining explicit boundary delimiters in its instructions.
Capability inventory: The skill possesses capabilities for command execution and local file writing (e.g., creating storage lifecycle policy JSON files).
Sanitization: The instructions recommend using Python's json module to parse and sort data, which facilitates safer processing of external content.

azure-cost