zype

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

• Third-party content exposure detected (high risk: 0.70). This skill instructs the agent to connect via the Membrane CLI to a Zype account and run actions (e.g., "membrane action run ... --connectionId=CONNECTION_ID") that fetch Zype data (videos/playlists/users), which is untrusted, user-generated third‑party content the agent is expected to read and could materially influence subsequent actions.