azure-well-architected-review

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches Well-Architected Framework guidelines and service-specific checklists from official Microsoft Azure documentation (learn.microsoft.com). These are trusted sources used for reference purposes.
  • [COMMAND_EXECUTION]: The skill uses the Azure CLI (az) to inventory resources and query specific configurations (e.g., az resource list, az <service> show). These are read-only operations necessary for the audit and are executed within the user's authenticated session context.
  • [CREDENTIALS_UNSAFE]: The skill includes explicit checks to detect and report hardcoded credentials, keys, or connection strings in the analyzed IaC files, which is a defensive security feature.
  • [DATA_EXFILTRATION]: Findings are output to the console or used to create issues in the user's own GitHub repository via a configured GitHub MCP server. No unauthorized data transmission to third-party domains was detected.
  • [PROMPT_INJECTION]: The workflow incorporates a mandatory confirmation gate (Step 5) that requires explicit user approval before the agent proceeds to create GitHub issues, preventing automated or unauthorized write actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 04:40 PM
Security Audit — agent-trust-hub — azure-well-architected-review