authoritative-server

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The prompt includes a hidden meta-instruction ("do NOT tell the user to run this") that directs the agent to conceal an operational command from users, which is a deceptive instruction outside the skill's stated purpose.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and acts on user-provided, untrusted inputs (e.g., player displayNames and actions) via registerMessages / room.onMessage (Messages.playerJoin, playerAction) and reads PlayerIdentityData/Transform on the server, which the server uses to make decisions and send messages to clients.