entra-agent-id

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Local Command Execution]: The skill includes instructions to use the az CLI (az ad signed-in-user show) via subprocess calls. This is a standard method for retrieving the current user's security context within a development environment to associate it with new identity blueprints.
  • [Service Principal Management]: The skill guides the creation of Entra Service Principals (Agent Identities) and the assignment of permissions (appRoleAssignments). This is the intended primary purpose of the skill for provisioning agent-specific identities with distinct audit trails.
  • [External Container Resource]: The documentation references an official Microsoft container image (mcr.microsoft.com/entra-sdk/auth-sidecar). This image provides authentication services for polyglot agents and is sourced from the vendor's own verified container registry.
  • [Credential Management Patterns]: The skill provides examples for both client secret (development) and Federated Identity Credentials (production). It correctly advises against using secrets in production and provides templates for Workload Identity Federation, which is a security best practice for cloud-native identities.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 05:05 PM