azure-advisor-report

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses authenticated Azure CLI commands to interact with official Microsoft Azure APIs (management.azure.com). No unauthorized network activity or exfiltration to third-party domains was identified.
  • [SAFE]: The skill includes a dedicated guardrail file (references/report-guardrails.md) that provides clear instructions to strip internal company metadata, credentials, and internal links from the final report.
  • [PROMPT_INJECTION]: The skill processes data from Azure Advisor and Cost Management APIs which is interpolated into reports. This creates a surface for indirect prompt injection if source data was manipulated, though the impact is low as the data is used for static document generation.
  • Ingestion points: advisor-*.json and cost-monthly.json files generated via Azure CLI (references/data-collection.md).
  • Boundary markers: No explicit delimiters are used in the Markdown or PPTX templates.
  • Capability inventory: File system access for writing reports and PowerShell execution for data processing.
  • Sanitization: No sanitization is performed on the data retrieved from Azure APIs prior to report inclusion.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 06:28 AM