The Agent Skills Directory

[SAFE]: The skill is a legitimate tool for interacting with Azure Data Explorer (Kusto). It focuses on query execution, schema discovery, and resource management within the Azure ecosystem.
[COMMAND_EXECUTION]: The skill provides fallback instructions using Azure CLI commands (e.g., az kusto cluster list, az rest). These commands are standard for managing Azure resources and are constructed using environment-specific variables like subscription IDs and cluster names.
[PROMPT_INJECTION]: The skill processes external log and telemetry data, creating an indirect prompt injection surface (Category 8). This is inherent to the skill's purpose of log analysis.
Ingestion points: Untrusted data enters the agent context via kusto_query tool output and az rest command results.
Boundary markers: No explicit markers or "ignore instructions" delimiters are specified for the processed log data.
Capability inventory: The skill possesses capabilities for listing Azure resources and executing arbitrary KQL queries via the kusto_query tool and shell commands.
Sanitization: No specific sanitization or filtering of the retrieved log content is described before the agent processes the information.

azure-kusto