deploying-to-azure

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a delegation pattern where a subagent processes output from az containerapp logs show to diagnose failures.
  • Ingestion points: External log data from Azure Container Apps is fed into the subagent's prompt in SKILL.md.
  • Boundary markers: The example prompt does not use specific delimiters to isolate untrusted log content from the agent's instructions.
  • Capability inventory: The subagent is tasked with running subsequent Azure CLI commands and suggesting fixes based on the logs.
  • Sanitization: There is no mention of sanitizing or validating the log output before analysis.
  • [COMMAND_EXECUTION]: The deployment workflow relies on local PowerShell scripts (preprovision.ps1, postprovision.ps1, predeploy.ps1) to automate environment setup, RBAC assignments, and Docker builds. These operations are standard components of the Azure Developer CLI workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 06:20 PM