The Agent Skills Directory

[Indirect Prompt Injection]: The skill processes untrusted telemetry data from Azure Monitor (e.g., AppEvents, AppExceptions) to generate KQL queries and Operations Agent instructions. This represents a potential vulnerability surface where external data could attempt to influence the agent's logic. The skill mitigates this by requiring mandatory schema verification and join-key validation against real data (Stage 12) before any correlation model is finalized.
[Command Execution]: The skill utilizes Azure CLI (az) commands for tasks such as resource discovery, identity verification, and API interactions. these operations are executed within the user's authenticated session and rely on the user's existing permissions.
[Privilege Escalation]: The instructions include a potential step to assign the 'Owner' role to a Fabric workspace identity on an Azure Log Analytics workspace. While this involves granting high-level permissions, the skill is designed to check for the user's authorization and requires explicit confirmation before performing the assignment.
[Dynamic Execution]: The skill dynamically generates and executes management commands, such as registering external Delta tables in a KQL database. These actions are performed to configure the environment based on discovered schema and are central to the skill's primary purpose.

azmon-mirroredcatalogs-operations-cli