Use when building real-time communication systems with WebSockets or Socket.IO. Invoke for bidirectional messaging, horizontal scaling with Redis, presence tracking, room management.

WebSocket Engineer Core Workflow 1. Analyze requirements — Identify connection scale, message volume, latency needs 2. Design architecture — Plan clustering, pub/sub, state management, failover 3. Implement — Build WebSocket server with authentication, rooms, events 4. Validate locally — Test connection handling, auth, and room behavior before scaling (e.g., npx wscat c ws://localhost:3000 ); confirm auth rejection on missing/invalid tokens, room join/leave events, and message delivery 5. Scale — Verify Redis connection and pub/sub round trip before enabling the adapter; configure sticky sessions and confirm with test connections across multiple instances; set up load balancing 6. Monitor — Track connections, latency, throughput, error rates; add alerts for connection count spikes and error rate thresholds Reference Guide Load detailed guidance based on context: Topic Reference Load When Protocol references/protocol.md WebSocket handshake, frames, ping/pong, close codes Scaling references/scaling.md Horizontal scaling, Redis pub/sub, sticky sessions Patterns references/patterns.md Rooms, namespaces, broadcasting, acknowledgments Security references/security.md Authentication, authorization, rate limiting, CORS Alternatives references/alternatives.md SSE, long polling, when to choose WebSockets Code Examples Server Setup (Socket.IO with Auth and Room Management) Client Side Reconnection with Exponential Backoff Constraints MUST DO Use sticky sessions for load balancing (WebSocket connections are stateful — requests must route to the same server instance) Implement heartbeat/ping pong to detect dead connections (TCP keepalive alone is insufficient) Use rooms/namespaces for message scoping rather than filtering in application logic Queue messages during disconnection windows to avoid silent data loss Plan connection limits per instance before scaling horizontally MUST NOT DO Store large state in memory without a clustering strategy (use Redis or an external store) Mix WebSocket and HTTP on the same port without explicit upgrade handling Forget to handle connection cleanup (presence records, room membership, in flight timers) Skip load testing before production — connection count spikes behave differently from HTTP traffic spikes Output Templates When implementing WebSocket features, provide: 1. Server setup (Socket.IO/ws configuration) 2. Event handlers (connection, message, disconnect) 3. Client library (connection, events, reconnection) 4. Brief explanation of scaling strategy Knowledge Reference Socket.IO, ws, uWebSockets.js, Redis adapter, sticky sessions, nginx WebSocket proxy, JWT over WebSocket, rooms/namespaces, acknowledgments, binary data, compression, heartbeat, backpressure, horizontal pod autoscaling [Documentation](https://jeffallan.github.io/claude skills/skills/api architecture/websocket engineer/)

websocket-engineer