azure-resource-visualizer
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill processes external metadata, including resource names, tags, and configuration properties retrieved via Azure CLI or Resource Graph queries in
SKILL.md. While this data is typically controlled by the resource owner, it represents an ingestion point for untrusted content if external parties can modify resource tags. The skill mitigates risk by explicitly instructing the agent to use placeholders for secrets and sensitive values during analysis. - Vendor-Specific Extension Management: The reference to
az extension add --name resource-graphinreferences/azure-resource-graph.mdinvolves the installation of an official Microsoft Azure CLI extension. This is a standard requirement for cross-subscription querying and follows the vendor's intended workflow for infrastructure discovery. - Sensitive Data Handling: The skill includes explicit directives in
SKILL.mdto identify and mask secrets, connection strings, and keys with placeholder names. This proactive instruction reduces the risk of accidental exposure of credentials within the generated diagrams or documentation.
Audit Metadata