game-demo-feedback-triage

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text is ingested when the operating user provides a “feedback packet” containing player/creator/streamer comments or quotes (e.g., Next Fest/Discord/Steam Playtest responses) and the skill uses that text to produce the LLM brief; this is indirect prompt injection risk via untrusted user-supplied prose.