jellyfin-cli

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow exposes the outsider’s free text via the search/recent command inputs (e.g., --query) that are sent to Jellyfin at runtime (/Search/Hints → cmd_search → client.search → JellyfinClient._get), so the agent will ingest whatever user text it’s given before selecting a specific item.