Redesigning a Routine That Failed Silently for a Week

A nightly issue-triage cloud routine has been running for weeks. You discover, by accident, that it hasn't posted to Slack in seven days — nobody noticed because the channel is quiet and low-traffic anyway, and a quiet channel looked normal. Diagnose the plausible causes given how routines actually work, and redesign the setup (prompt, monitoring, or both) so this specific failure mode can't recur silently.