microsoft365-expert
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates processing external data from the Microsoft 365 ecosystem, creating an attack surface for malicious instructions.
- Ingestion points: references/EXAMPLES.md provides methods to read email content (listEmails) and SharePoint files (Get_file_content).
- Boundary markers: There are no specific delimiters or instructions to ignore instructions embedded in the external data.
- Capability inventory: The skill enables high-privilege actions via the Graph API, such as sending emails and posting to Teams, and includes access to the Bash tool.
- Sanitization: The provided implementation patterns do not include logic for sanitizing or escaping retrieved content before use in further operations.
- [COMMAND_EXECUTION]: The skill requests the Bash tool for running administrative PowerShell scripts and CLI utilities. This provides a capability that could be misused if untrusted input is passed to shell commands without proper validation.
Audit Metadata