Status endpoints

Snapshots, summaries, monitors and uptime, with full payload shapes.

All four are public and unauthenticated. They read the same cached snapshot, so a page render and an API call can never disagree about the current state.

`GET /status`

The full public snapshot — everything the page renders.

bash

curl https://status.acme.com/api/heyo-status/status

json

{  "generatedAt": "2026-01-01T12:00:00.000Z",  "retentionDays": 90,  "displayDays": 90,  "overall": "degraded",  "groups": [    { "name": "Web", "status": "operational", "monitors": ["Website", "Dashboard"] },    { "name": "API", "status": "degraded", "monitors": ["API", "Webhooks"] }  ],  "monitors": [    {      "name": "API",      "type": "http",      "description": "Public REST API.",      "group": "API",      "latestStatus": false,      "latestCheckedAt": "2026-01-01T11:55:00.000Z",      "latestLatencyMs": 142,      "uptimePercent": 99.82,      "uptimePercent24h": 97.2,      "uptimePercent7d": 99.6,      "uptimePercent30d": 99.8,      "uptimePercent90d": 99.82,      "avgLatencyMs": 118,      "days": [        {          "key": "2025-10-04",          "label": "Oct 4, 2025",          "status": "up",          "tooltip": "100% uptime — 288 checks",          "totalChecks": 288,          "downChecks": 0,          "downMinutes": 0,          "downPercent": 0        }      ],      "latencySeries": [        {          "key": "2026-01-01T11:00:00.000Z",          "label": "Jan 1, 2026, 11:00 UTC",          "at": "2026-01-01T11:00:00.000Z",          "latencyMs": 121,          "totalChecks": 12        }      ]    }  ],  "activeIncidents": [],  "scheduledMaintenances": [],  "hasConfig": true,  "hasData": true}

overall"operational" | "degraded" | "major-outage" | "maintenance"optional

Maintenance wins, then a critical incident, then any incident, then the monitors themselves.

daysDaySnapshot[]optional

Exactly displayDays entries, oldest first. status is up, down, degraded or empty.

latencySeriesLatencySample[]optional

Hourly averages over the last 24 hours while raw rows exist, otherwise daily averages over 7 days. Empty buckets are omitted.

hasConfig / hasDatabooleanoptional

Whether any public monitor is configured, and whether any check has landed. Themes use them for the onboarding state.

Private monitors never appear here.

`GET /status/summary`

A deliberately small payload for embeds, uptime widgets and dashboards — no day grid, no latency series.

bash

curl https://status.acme.com/api/heyo-status/status/summary

json

{  "generatedAt": "2026-01-01T12:00:00.000Z",  "overall": "operational",  "hasConfig": true,  "hasData": true,  "activeIncidents": 0,  "scheduledMaintenances": 1,  "groups": [{ "name": "Web", "status": "operational", "monitors": ["Website"] }],  "monitors": [    {      "name": "Website",      "type": "http",      "group": "Web",      "latestStatus": true,      "latestCheckedAt": "2026-01-01T11:55:00.000Z",      "latestLatencyMs": 96,      "uptimePercent": 99.98    }  ]}

activeIncidents and scheduledMaintenances are counts here, not arrays.

`GET /monitors`

The monitors array from the snapshot, unchanged.

bash

curl https://status.acme.com/api/heyo-status/monitors

`GET /monitors/:name`

One monitor, or 404 { "error": "Monitor not found." }. Percent-encode names with spaces.

bash

curl https://status.acme.com/api/heyo-status/monitors/APIcurl https://status.acme.com/api/heyo-status/monitors/Web%20App

`GET /monitors/:name/uptime`

Just the numbers, for an SLA report or a dashboard tile.

bash

curl https://status.acme.com/api/heyo-status/monitors/API/uptime

json

{  "name": "API",  "24h": 97.2,  "7d": 99.6,  "30d": 99.8,  "90d": 99.82,  "window": 99.82,  "displayDays": 90,  "avgLatencyMs": 118}

window is uptime across the configured display window, which is not always 90 days — displayDays tells you what it covers. A null means no checks exist for that period, which is different from 0.

`GET /healthz`

json

{ "ok": true }

It answers whether the status API itself is serving — it does not read the database and does not reflect monitor state. Use it for a load-balancer probe; use /status/summary for actual service health.

Caching

Snapshot reads are served from the in-process cache for cache.snapshotTtlMs (default 30 s). Check runs and admin mutations invalidate it. Responses carry no Cache-Control, so add your own edge caching if you need it — badges are the exception and set public, max-age=60 themselves.

GET /status

GET /status/summary

GET /monitors

GET /monitors/:name

GET /monitors/:name/uptime

GET /healthz

Caching

Related

`GET /status`

`GET /status/summary`

`GET /monitors`

`GET /monitors/:name`

`GET /monitors/:name/uptime`

`GET /healthz`