gamekit

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill content is educational and documentation-focused, providing Swift code samples for GameKit features. No malicious code, obfuscation, or unauthorized data access patterns were detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill documentation correctly identifies the potential for indirect injection via multiplayer match data and provides mitigation guidance.
  • Ingestion points: Data received from remote players via GKMatchDelegate in SKILL.md and GKTurnBasedEventListener in references/gamekit-patterns.md.
  • Boundary markers: Not applicable to this documentation-style skill, which focuses on providing implementation code for the agent to follow.
  • Capability inventory: The skill describes capabilities such as saving data to iCloud (saveGameData) and submitting scores to leaderboards.
  • Sanitization: The skill explicitly warns to 'Treat received match data as untrusted input', which is a best-practice security recommendation for handling data from remote players.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:59 AM
Security Audit — agent-trust-hub — gamekit