game-narrative-director

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The instructions establish a professional persona for narrative design without suspicious command execution, network exfiltration, or credential access.\n- [INDIRECT_PROMPT_INJECTION]: The skill explicitly discusses the attack surface of processing player-controlled natural language in a game environment.\n
  • Ingestion points: The documentation describes real-time player input for AI NPC interaction systems (Inworld AI, Convai).\n
  • Boundary markers: Recommends the use of authored narrative spines to constrain AI output and pre-defined topic guardrails.\n
  • Capability inventory: The skill specifies standard file system tools (Read, Write, Edit) for managing narrative assets.\n
  • Sanitization: Recommends multi-stage filtering, including profanity filters, topic guardrails, and adversarial testing to harden against player manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 11:53 AM