azure-virtual-desktop
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill is designed to fetch content from the official Microsoft Learn domain (
learn.microsoft.com). This is a legitimate function for providing the agent with the latest technical documentation. - [INDIRECT_PROMPT_INJECTION]: By design, the skill ingests external data into the agent's context. While this introduces a surface for indirect prompt injection, the risk is mitigated by the use of trusted, official documentation sources.
- [COMMAND_EXECUTION]: The instructions reference the
mcp_microsoftdocstool and its installation guide on GitHub. This is a vendor-provided tool designed to integrate Microsoft documentation with AI agents and is hosted on the organization's official repository.
Audit Metadata