skills/ssheleg/xr-dev/quest-perf/Gen Agent Trust Hub

quest-perf

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes commands through the metavr CLI and adb to manage headset applications, capture system logs, and control performance profiling sessions.\n- [EXTERNAL_DOWNLOADS]: Facilitates the installation of essential profiling tools from trusted sources, including the Meta fork of RenderDoc and the OVR Metrics Tool, via the metavr tools install command.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the device environment through system logs and performance traces (ingestion points in SKILL.md and references/capture-playbook.md). It possesses the capability to execute shell commands via metavr. While boundary markers and sanitization are not explicitly defined for these inputs, the risk is considered low due to the technical nature of the logs and existing agent guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 06:45 AM
Security Audit — agent-trust-hub — quest-perf