tabletopkit

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The instructions and code samples focus exclusively on the legitimate use of Apple's TabletopKit framework for game development on visionOS.\n- [EXTERNAL_DOWNLOADS]: The skill includes documentation links to a third-party mirror (sosumi.ai) for Apple developer resources. These are static reference links and do not involve automated script execution or downloading untrusted code into the agent's environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill manages multiplayer game state, which is an ingestion point for external data.\n
  • Ingestion points: TabletopAction and TableSnapshot data processed in GameObserver implementation (references/tabletopkit-patterns.md).\n
  • Boundary markers: Instructions emphasize that validation and turn ownership must be explicit (SKILL.md).\n
  • Capability inventory: Capabilities are restricted to RealityKit rendering and TabletopKit state synchronization; no access to sensitive system commands or file systems.\n
  • Sanitization: Implementation of validateAction and deterministic apply logic serves as a sanitization mechanism for incoming actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:14 AM
Security Audit — agent-trust-hub — tabletopkit