azure-databricks

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly requires fetching external Markdown docs at runtime (e.g., https://learn.microsoft.com/en-us/azure/databricks/admin/account-settings/audit-logs) using mcp_microsoftdocs or fetch_webpage and injects that content into the agent context, meaning remote content can directly control prompts/answers.