infinite-pokemon-npc

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill bundle contains no executable scripts (Python, JavaScript, Shell) or binaries. It is composed entirely of markdown instructions, a license, and reference documentation.
  • [EXTERNAL_DOWNLOADS]: The documentation includes informational links to the author's official GitHub repository (github.com/Shellishack/infinite-pokemon) for project documentation and source code provenance. These are static references and do not trigger automated downloads or code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest game state observations to generate NPC dialogue and behavior proposals.
  • Ingestion points: External data is supplied via a "saved-run observation" or "map-design context" referenced in SKILL.md.
  • Boundary markers: The skill instructs the agent to "read the supplied observation once" and restricts behavior to the "behavior contract" defined in references/behavior.md.
  • Capability inventory: The skill has no capability to execute commands, write to the file system, or perform network operations; it only produces structured JSON data.
  • Sanitization: The instructions explicitly prohibit the AI from creating player achievements or inventing game buffs based on input data, and state that only the engine (not the AI) resolves game logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 01:31 AM
Security Audit — agent-trust-hub — infinite-pokemon-npc