hz-xr-simulator-control

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the metavr and hzdb CLI tools to interact with the XR simulator. Commands include runtime logs, runtime list, runtime status, app launch, app quit, device set, etc. All tools are within the vendor's namespace.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads output from the XR simulator (logs, runtime status, list of runtimes), creating a potential surface for indirect injection from the environment being tested.
  • Ingestion points: Data retrieved via metavr xrsim runtime logs, metavr xrsim runtime list, and metavr xrsim runtime status.
  • Boundary markers: The skill instructs the agent to use --format json for parsing, providing structure but not a security boundary.
  • Capability inventory: Capabilities are limited to the metavr and hzdb command groups via Bash.
  • Sanitization: There is no mention of filtering or sanitizing the content of simulator logs or status output.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 06:28 PM
Security Audit — agent-trust-hub — hz-xr-simulator-control