azure-resource-lookup

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Indirect Prompt Injection Surface: The skill translates natural language requests into Azure Resource Graph queries. This creates a surface where untrusted user input influences command construction.
  • Ingestion points: Natural language user requests in SKILL.md processed by extension_cli_generate to build CLI commands.
  • Boundary markers: The instructions do not specify explicit delimiters for the generated query strings within the CLI command.
  • Capability inventory: The skill utilizes az graph query for read-only metadata retrieval and az extension add for configuring the environment.
  • Sanitization: The skill relies on the underlying tool's generation logic and the inherently read-only nature of the Azure Resource Graph service.
  • Dynamic Command Generation: The skill identifies user intent and constructs specific Azure CLI commands at runtime based on that intent.
  • Evidence: The workflow described in SKILL.md involves generating and executing commands like az graph query -q "<KQL>".
  • External Extension Management: The skill provides instructions to install necessary components for the Azure CLI to support resource graph queries.
  • Evidence: SKILL.md references the execution of az extension add --name resource-graph to ensure the local environment has the required capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:06 AM
Security Audit — agent-trust-hub — azure-resource-lookup