websocket-engineer

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md "Workflow 1: Scalable Socket.IO Server" shows the server directly ingesting and acting on arbitrary client messages (e.g., socket.on("message", (data) => { io.to(data.room).emit("chat", data.text); }) and socket.on("join", (room) => socket.join(room))), which clearly consumes untrusted/user-generated third‑party content from clients and uses it to drive runtime actions.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The prompt explicitly instructs modifying system-wide settings (e.g., editing /etc/security/limits.conf and running sysctl -w) which require elevated privileges and change the machine state, so it encourages actions that can compromise the host.