skills/melvynx/cli-skills/az/Gen Agent Trust Hub

az

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the official Azure Command Line Interface using the brew install azure-cli command. This is a well-known tool from a reputable source.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of the az binary for cloud resource management, including commands for authentication, resource groups, virtual machines, and Kubernetes clusters.
  • [PROMPT_INJECTION]: The skill documentation includes examples where user-provided inputs like resource names and file paths are used as command arguments, creating a surface for indirect prompt injection.
  • Ingestion points: User-supplied strings for resource identifiers (--name, --resource-group) and local file paths (--src-path, --file) in the SKILL.md examples.
  • Boundary markers: None provided in the command templates to delimit user input from shell arguments.
  • Capability inventory: Includes wide-ranging management capabilities such as resource deletion, VM management, and Kubernetes credential retrieval via the az CLI as seen in SKILL.md.
  • Sanitization: The instructions do not specify input validation or sanitization for these parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 04:14 AM
Security Audit — agent-trust-hub — az