azure-information-protection

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly requires fetching live Microsoft Docs pages at runtime (e.g., https://learn.microsoft.com/en-us/azure/information-protection/monitor-rms-connector) via mcp_microsoftdocs:microsoft_docs_fetch or fetch_webpage, and that fetched Markdown is injected into the agent context so it can directly control prompts.