ms365-tenant-manager
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external CSV files (e.g., user lists) and interpolates their contents into generated PowerShell scripts without sanitization.
- Ingestion points:
scripts/user_management.pyandscripts/powershell_generator.pyread user-supplied CSV data via the--csvargument. - Boundary markers: The generated scripts do not include explicit boundaries or warnings to ignore instructions embedded in the data fields.
- Capability inventory: The skill generates scripts for high-privilege operations including user creation, license assignment, and Conditional Access policy configuration.
- Sanitization: Input from CSV files (such as UserPrincipalName or DisplayName) is inserted directly into PowerShell string literals, which could lead to command injection in the generated scripts if the CSV content contains maliciously crafted characters.
- [DYNAMIC_EXECUTION]: The Python tools (
scripts/powershell_generator.py,scripts/user_management.py, andscripts/tenant_setup.py) dynamically assemble PowerShell code from predefined templates and user input. This script generation is a core feature but requires users to carefully review the output before execution. - [COMMAND_EXECUTION]: The skill is designed to produce scripts that require Global Administrator or similar high-level privileges for Microsoft 365 tenant management. Execution of these scripts without thorough review could lead to unintended configuration changes.
- [EXTERNAL_DOWNLOADS]: The skill documentation correctly identifies and recommends the installation of official Microsoft PowerShell modules (
Microsoft.Graph,ExchangeOnlineManagement,MicrosoftTeams) from Microsoft's official repositories.
Audit Metadata