NYC

azure-expert

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill requests permission for Bash(az:*). While powerful, this is the primary intended function of an Azure Expert skill and is used for resource management.
  • [CREDENTIALS_UNSAFE] (SAFE): Example code includes placeholders like YOUR_KEY and generic endpoints, following safe documentation practices.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The Azure Function examples demonstrate ingesting data from external triggers (HTTP/Queue), which represents a standard but note-worthy attack surface. Ingestion points: req.get_json() and msg.get_body(). Boundary markers: None. Capability inventory: Full Azure CLI (az) access. Sanitization: Not demonstrated in the simplified examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:01 PM