Seiri is push-based ("heartbeat") monitoring for cron jobs, background workers, and Kubernetes CronJobs — not website uptime monitoring, and not a cron-to-webhook scheduler. The model is inverted from typical uptime tools: your job pings Seiri when it finishes, and if that ping doesn't arrive within the expected window, Seiri alerts you immediately — catching the silent failures that never throw an error and never show up in Datadog, Grafana, or CloudWatch.
Every monitor gets a unique ping URL and a unique @ping.seiri.app email address, so jobs running on locked-down networks or legacy servers that can send mail but not make outbound HTTP calls can still check in.
Kubernetes is a first-class surface, not a bolt-on. The seiri-kube-agent Helm operator installs into a cluster and watches HealthCheck custom resources, continuously checking Deployments, Pods, Jobs, StatefulSets, and DaemonSets, reporting phases (Healthy, Unhealthy, Missing) back to Seiri.
Other capabilities: multi-channel alerting (Slack, Discord, email, SMS, custom webhooks) with escalation, public or custom-domain status pages, a REST API with scoped API keys, and a free-forever tier (10 monitors, 5-minute check interval, 3 seats).