NYC

azure-resource-lookup

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill documentation recommends installing the 'resource-graph' Azure CLI extension using az extension add. This is an official extension provided by Microsoft, qualifying it as a trusted source per [TRUST-SCOPE-RULE].\n- [COMMAND_EXECUTION] (SAFE): Usage is restricted to az graph query, a read-only KQL command designed for retrieving indexed resource metadata. It does not possess capabilities for resource modification or deletion.\n- [PROMPT_INJECTION] (LOW): The skill creates an indirect prompt injection surface by interpolating user natural language intent into generated KQL queries without explicit boundary markers or sanitization logic.\n
  • Ingestion points: User intent is ingested via the extension_cli_generate tool to produce commands.\n
  • Boundary markers: No delimiters (like triple backticks) or 'ignore embedded instructions' warnings are specified in the prompt interpolation instructions.\n
  • Capability inventory: The skill executes CLI commands via az (subprocess) and accesses resource metadata.\n
  • Sanitization: There is no evidence of validation or sanitization of the generated query string before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 08:28 PM