azure-resource-lookup

Pass

Audited by Gen Agent Trust Hub on Oct 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution: The skill generates and executes az graph query commands to retrieve metadata about Azure resources. These operations are conducted using the Azure CLI within the user's authenticated environment, allowing for cross-subscription resource inventory and management.
  • Indirect Prompt Injection Surface: The skill translates user intents into KQL queries, representing a surface where untrusted data could influence agent actions. 1. Ingestion points: User requests processed via the intent parameter in extension_cli_generate (SKILL.md). 2. Boundary markers: Explicit boundary markers are not defined within the skill files; the skill depends on the platform's command generation logic. 3. Capability inventory: Execution of Azure CLI commands (az graph query) and subscription/group listing tools. 4. Sanitization: The skill instructions do not implement custom sanitization, relying instead on the generated KQL syntax for structural execution.
  • External Extension Management: The instructions include the installation of the resource-graph extension for the Azure CLI. This is a standard, vendor-provided extension required for the skill's primary functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 2, 2026, 09:01 AM