azure-automation

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly requires runtime fetching of external Markdown documentation (e.g., via mcp_microsoftdocs:microsoft_docs_fetch or fetch_webpage) such as https://learn.microsoft.com/en-us/azure/automation/troubleshoot/collect-data-microsoft-azure-automation-case, which will be injected into the agent context and therefore can directly control prompts/instructions at runtime.