production-readiness

Makes code behave well in production, in any language, through logging, metrics, tracing, timeouts, retries, graceful failure and safe releases. Use whenever code calls a network or another service, runs as a job or a queue consumer, needs logs or alerts, can fail partially, or is about to ship, even if the user only says "agregá logs" or "esto se cae a veces". Covers circuit breakers, health checks, feature flags, rollbacks and "qué logueo", "cómo reintento", "está listo para producción" and "cómo lo despliego sin romper". For finding the cause of a specific failure use debugging; for designing deployment topology use the architect agent.