microsoft-code-reference
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation provides instructions for using the official
@microsoft/learn-clipackage vianpxor globalnpminstallation as a fallback mechanism when the primary API is unavailable. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates fetching external content from
learn.microsoft.com. While this introduces a surface where untrusted data could enter the agent's context, the target source is a well-known and trusted official documentation service. - Ingestion points: Content is retrieved via the
microsoft_docs_fetchtool from external URLs. - Boundary markers: Not explicitly defined in the provided snippets.
- Capability inventory: The skill allows for documentation lookups and code sample searches; it does not demonstrate capabilities for arbitrary shell execution on the fetched data.
- Sanitization: None specified for the fetched documentation text.
- [COMMAND_EXECUTION]: The documentation includes example shell commands for using the
mslearnCLI tool. These commands are intended for manual user execution or as a reference for how the agent should interact with the tool if installed.
Audit Metadata