create-game

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a project generator. It copies a local boilerplate folder to a new project directory and provides templates for game planning.
  • [COMMAND_EXECUTION]: The skill uses standard shell commands (cp -R, npm install) to initialize the project environment. These are restricted to the local filesystem and established package registries (npm).
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill uses localStorage via a useStorage hook to save and load game state. This is a standard practice for web applications and does not involve exfiltrating sensitive user data to external servers.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user input for the game name and rule variations. While this represents an attack surface, the instructions include specific parsing steps and conversational checkpoints to ensure the agent maintains its primary role as a game developer.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 09:19 PM
Security Audit — agent-trust-hub — create-game