azure-upgrade

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Infrastructure Management Commands: The skill utilizes various Azure CLI commands to query and modify cloud resources. These operations typically require elevated permissions (Owner or Contributor roles) within an Azure subscription to perform tasks such as resource creation and configuration migration.
  • External Utility Dependencies: The workflow depends on the resource-graph Azure CLI extension and the jq command-line JSON processor. These are well-recognized tools in the Azure ecosystem used for querying resource metadata and processing command output.
  • Handling of Sensitive Configurations: The skill retrieves application settings and managed identity details from existing Azure services. While this data is handled within the context of the migration, users should be aware that these configurations often contain environment-specific settings.
  • Retrieval of Deployment Artifacts: There are procedures for downloading function app deployment packages from Azure Storage blobs. This is intended for scenarios where source code is not readily available in a repository.
  • Automated Command Generation: The skill generates executable scripts based on environment discovery. These scripts are intended to be idempotent and follow a step-by-step validation process, which is a standard safety practice for infrastructure automation.
  • Indirect Prompt Injection Surface: The skill processes external data from Azure resource configurations to generate assessment reports.
  • Ingestion points: Commands in references/services/functions/automation.md ingest data from the Azure environment metadata.
  • Boundary markers: Output is formatted into Markdown reports in references/services/functions/assessment.md using standard delimiters.
  • Capability inventory: The skill uses az CLI commands and curl for resource management and validation.
  • Sanitization: Ingested metadata is used for technical assessment; users should review reports before proceeding with upgrades.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 08:52 PM