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

quest-native

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill serves as a legitimate technical guide for VR development on Horizon OS. It directs agents to official documentation hosted by Meta (developers.meta.com) and refers to industry-standard libraries such as the Khronos OpenXR loader. All referenced code samples and build configurations follow standard Android and NDK development practices.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface by instructing the agent to search external documentation and analyze local project files.
  • Ingestion points: Usage of the metavr tool for searching Meta's official developer documentation and reading local project files such as AndroidManifest.xml and build.gradle (specified in SKILL.md and references/manifest-and-gradle.md).
  • Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore potentially malicious content embedded in documentation search results or local project files.
  • Capability inventory: The agent is empowered to provide architectural advice, manifest corrections, and build script modifications based on the ingested data.
  • Sanitization: No explicit sanitization or validation logic is defined for data retrieved from external documentation tools or local project files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 06:45 AM
Security Audit — agent-trust-hub — quest-native