aca-sandboxes

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • CLI Installation Scenarios: The skill documents how to install the aca CLI using one-liner scripts from Microsoft's aka.ms service. While these scripts are executed directly in the shell, they originate from the official vendor domain and are standard for this utility's deployment.
  • Controlled Remote Execution: The primary function of the skill is to allow users and agents to execute commands within Azure sandboxes. The use of microVM isolation provides a robust security boundary, and the skill includes instructions for managing these sessions securely via the aca sandbox exec and shell commands.
  • Egress and Network Security: It highlights the ability to restrict outbound network traffic from sandboxes using a 'deny-default' egress policy. This is a critical security consideration for isolating workloads and preventing unauthorized communication.
  • Identity Integration: The skill correctly utilizes Azure's managed identities and Entra ID for authentication, leveraging the existing security infrastructure of the Azure platform without requiring the management of long-lived secrets within the skill itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 05:06 PM
Security Audit — agent-trust-hub — aca-sandboxes