The Agent Skills Directory

Indirect Prompt Injection Surface: The skill processes data returned from external Kusto databases, creating a potential surface where malicious instructions embedded in database records could be interpreted by the agent.
Ingestion points: Data retrieved via kusto_query as described in SKILL.md enters the agent's context.
Boundary markers: The skill does not currently specify delimiters or instructions for the agent to ignore instructions within the retrieved data.
Capability inventory: The skill has access to database query tools and shell-based Azure CLI fallback operations.
Sanitization: There is no mention of sanitizing or validating the retrieved data before it is interpreted.
Command Execution Considerations: The skill provides a fallback mechanism using az kusto and az rest commands. While these are official Microsoft tools for resource management, their use involves the generation of shell commands, which requires careful handling of user-supplied parameters to prevent potential command injection.

azure-kusto