azure-resource-lookup

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFE
Full Analysis
  • Standard Azure CLI Integration: The skill utilizes the az command-line interface and the official resource-graph extension. These are standard, well-documented tools for Azure administration provided by the vendor.\n- Read-Only Scope: The instructions specifically restrict the skill to resource discovery and health checks, explicitly prohibiting any deployment or modification of resources. This helps maintain a secure posture and prevents unintended changes to the infrastructure.\n- Controlled Command Generation: By using extension_cli_generate, the skill structures the way KQL queries are built, ensuring they are focused on specific resource lookup tasks rather than arbitrary execution.\n- Resource Metadata Ingestion Surface: The skill processes external Azure resource data like names and tags. \n
  • Ingestion points: External metadata enters the context via az graph query results in the azure-resource-lookup skill.\n
  • Boundary markers: No specific delimiters are used to separate resource data from instructions in the prompt construction.\n
  • Capability inventory: The skill uses az graph query and extension_cli_generate across its workflow.\n
  • Sanitization: No explicit sanitization or filtering of resource metadata is performed before the data is processed by the agent.\nWhile this represents a potential surface for indirect instruction, the risk is minimal given the skill's administrative purpose and the agent's constrained usage of the output.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 03:54 AM