API Reference

Routes

All routes are relative to the mount point (e.g., /monitoring).

Method	Path	Description
GET	/api/status	Overall status with all check states
GET	/api/checks	List all checks
GET	/api/checks/:id	Get single check
POST	/api/checks	Create check
PUT	/api/checks/:id	Update check (partial)
DELETE	/api/checks/:id	Delete check
POST	/api/checks/:id/toggle	Enable/disable check
POST	/api/checks/:id/run	Run check immediately
GET	/api/incidents	List incidents (?check_id, ?status, ?limit)
GET	/api/alert-rules	List alert rules
POST	/api/alert-rules	Create alert rule
DELETE	/api/alert-rules/:id	Delete alert rule
GET	/api/maintenance	List maintenance windows
POST	/api/maintenance	Create maintenance window
DELETE	/api/maintenance/:id	Delete maintenance window
GET	/api/config	Export full config (checks + alert rules)
PUT	/api/config	Import checks (declarative sync)

State Machine

unknown  → healthy    (first success, no alert)
unknown  → degraded   (first failure, threshold not met)
healthy  → degraded   (failure, threshold not met)
degraded → unhealthy  (threshold met → onAlert 'failure')
unhealthy → healthy   (success → onAlert 'recovery')
unhealthy → unhealthy (still failing, no alert)

Alert Payload

When a state transition triggers an alert, onAlert receives:

interface AlertPayload {
  type: "failure" | "recovery";
  check: CheckConfig;
  checkState: CheckState;
  result: CheckResult;
  timestamp: string;
}

Dashboard

The embedded dashboard is served at the mount point root (e.g., /monitoring/). It includes:

• Overall system status
• Per-check status timeline
• Response time sparkline
• Auto-refresh every 60 seconds