Skip to main content
Version: 2.0.1 (preview)

Test-MtAIAgentEmailExfiltration

SYNOPSISโ€‹

Tests if AI agents can send email with AI-controlled inputs.

SYNTAXโ€‹

Test-MtAIAgentEmailExfiltration [-ProgressAction <ActionPreference>] [<CommonParameters>]

DESCRIPTIONโ€‹

Checks all Copilot Studio agents for email-sending tools (such as Office 365 Outlook or SendMail connectors) where the recipient, subject, or body may be controlled by AI-generated content. This presents a risk of data exfiltration via email to attacker-controlled addresses.

EXAMPLESโ€‹

EXAMPLE 1โ€‹

Test-MtAIAgentEmailExfiltration

PARAMETERSโ€‹

-ProgressActionโ€‹

{{ Fill ProgressAction Description }}

Type: ActionPreference
Parameter Sets: (All)
Aliases: proga

Required: False
Position: Named
Default value: None
Accept pipeline input: False
Accept wildcard characters: False

CommonParametersโ€‹

This cmdlet supports the common parameters: -Debug, -ErrorAction, -ErrorVariable, -InformationAction, -InformationVariable, -OutVariable, -OutBuffer, -PipelineVariable, -Verbose, -WarningAction, and -WarningVariable. For more information, see about_CommonParameters.

INPUTSโ€‹

OUTPUTSโ€‹

[bool] - Returns $true if no email exfiltration risk is found, $false if anyโ€‹

agent has email-sending capabilities with dynamic inputs, $null if data isโ€‹

unavailable.โ€‹

NOTESโ€‹

https://maester.dev/docs/commands/Test-MtAIAgentEmailExfiltration