One filtered, sorted, paged view of every monitor on the account: HTTP checks, heartbeats and browser checks together.

GET

/monitors

const url = 'https://api.siteqwality.com/monitors';
const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}

curl --request GET \
  --url https://api.siteqwality.com/monitors \
  --header 'Authorization: Bearer <token>'

total is the size of the whole filtered set, not the length of items.

This exists because the three monitor types cannot be paged separately and merged in the client. GET /monitors/list applies limit and offset to http_jobs and cron_jobs independently, so limit=10 can return twenty rows and one offset means two different things; browser checks were not in that endpoint at all and were merged in from an unfiltered list. A client asking for “the next page” of the combined array therefore asked for an offset that existed in none of the three sources, got nothing back, and concluded there was no more data while monitors it had never shown sat in the gap.

/monitors/list is unchanged: it is a documented public API surface.

Authorizations

bearer_auth

Parameters

Query Parameters

monitor_type

string | null

http, browser or cron. Blank values select all types.

limit

integer | null format: int64

Page size. Clamped to 1..=100.

offset

integer | null format: int64

Row offset into the filtered result set.

text

string | null

Free-text search over name and URI.

status

string | null

success, failed or pending_failure.

state

string | null

active or paused.

sort_field

string | null

Column to sort by. Unrecognized values fall back to the default order.

sort_order

string | null

asc or desc. Ignored without sort_field.

tags

Array<string>

Repeatable. A monitor matches when it carries every requested tag.

Responses

200

One page of matching monitors

application/json

One server-filtered, server-sorted page of the unified monitor list.

object

items

required

Array<object>

One row of the unified monitor list.

Fields are emitted as explicit null rather than omitted when they do not apply to the row’s type. The dashboard reads several of them positionally (an absent method and a null method are not the same thing to it), and bulk edit spreads the whole row back into an update request, so a missing key is a silently blanked setting rather than a cosmetic difference.

object

required

string format: uuid

monitor_type

required

http, cron or browser.

string

Allowed values: http cron browser

friendly_name

Display name. Browser checks store this as name; it is projected here so one column renderer serves all three types.

string | null

current_status

required

success, failed or (HTTP and browser only) pending_failure.

string

state

required

active or paused.

string

not_checked_reason

One of:

null
string

null

Set on an ACTIVE monitor that is not being checked: quota while the account’s quota for its product is used up. Show this instead of current_status.

string

Allowed values: quota

created_at

required

string format: date-time

last_status_change_at

string | null format: date-time

uptime_24h

Uptime percentage over the last 24h. HTTP monitors only, and null when the monitor has no checks in the window.

string | null

group_notifications

Notification groups, resolved in one batched query for the whole page.

array | null

object

required

string format: uuid

account_id

required

string format: uuid

friendly_name

required

string

delay_send_after_minutes

integer | null format: int32

resend_every_minutes

integer | null format: int32

created_at

required

string format: date-time

notifications

required

Array<object>

object

required

string format: uuid

account_id

required

string format: uuid

type

required

string

Allowed values: email sms slack webhook telegram discord microsoft_teams pagerduty opsgenie google_chat mattermost pushover pushbullet

details

required

One of:

object

email_address

required

string

object

pagerduty_integration_id

required

string format: uuid

pagerduty_severity

Severity of trigger events sent through this notification. Defaults to critical when the caller omits it.

string

Allowed values: critical error warning info

created_at

required

string format: date-time

health

One of:

null
object

null

How this channel’s recent deliveries went: whether the last alert reached anyone, and why not when it did not (a revoked Slack token, a bouncing inbox). Present on the read endpoints; absent on create and update responses.

object

state

required

healthy: the most recent delivery reached the provider. degraded: it was refused, bounced, complained about, could not be sent, or skipped for a reason the channel must fix (such as sms_quota), so the next alert will likely not arrive either. unknown: nothing sent in the lookback window. A skip by design, such as heartbeat_channel_unsupported, is listed in recent but never judged.

string

recent

required

Newest first, at most [CHANNEL_HEALTH_RECENT].

Array<object>

One delivery, as a channel’s health shows it.

object

required

string format: date-time

status

required

string

reason

The provider’s reason for a non-success status, URLs reduced to hosts. For skipped, a slug such as sms_quota or tls_channel_unsupported.

string | null

numb_of_sites

integer | null format: int64

service_id

string | null format: uuid

tags

array | null

last_run_at

string | null format: date-time

check_regions

array | null

min_healthy_regions

integer | null format: int32

timeout_ms

Milliseconds. Browser checks store seconds; the value is converted so the Timeout column reads true for them instead of blanking.

integer | null format: int32

uri

string | null

method

Explicitly null for heartbeats and browser checks. The dashboard’s legacy “no method means heartbeat” heuristic still exists as a fallback, so this must be present and null rather than omitted.

string | null

run_interval_seconds

integer | null format: int32

run_interval_thirty_seconds

boolean | null

follow_redirects

boolean | null

final_uri

string | null

keyword_search

string | null

keyword_search_inverted

boolean | null

keyword_search_case_sensitive

boolean | null

keyword_search_regex

boolean | null

request_body

string | null

content_type

One of:

null
string

null

auth_type

One of:

null
string

null

auth_username

string | null

has_auth

True when a basic-auth password or bearer token is stored for this monitor. The credentials themselves are write-only and are deliberately absent from this struct, so there is nothing here for a read-only viewer, a scoped API key, an M2M client or a read-only impersonation session to read back. See crate::model::api::secrets.

Bulk edit spreads a row back into an update request, and the update statement writes COALESCE($n, auth_password), so an absent key leaves the stored credential alone. That is why these two can be dropped from the shape outright rather than emitted as null.

boolean

success_http_codes

array | null

failure_http_codes

array | null

custom_http_headers

Names and allowlisted metadata values; other values become secrets::REDACTED. Sending the placeholder back on an update keeps the stored value.

object | null

tls_job_id

string | null format: uuid

monitor_tls

required

boolean

dns_job_id

string | null format: uuid

monitor_dns

required

boolean

page_speed_job_id

string | null format: uuid

monitor_page_speed

required

boolean

http_version

One of:

null
string

null

Preferred HTTP version for the check transport (isahc/libcurl). Stored on the job; observed negotiated version is on http_job_results.http_version.

string

Allowed values: auto http1_1 http2 http2_with_fallback

confirmation_threshold

integer | null format: int32

check_interval_seconds

integer | null format: int32

last_received_at

string | null format: date-time

last_checked_at

string | null format: date-time

frequency

Browser check cadence label (5m, 10m, 15m, 30m, 1h).

string | null

timeout_seconds

integer | null format: int32

total

required

Number of monitors matching the filters, across every page. This is what “X-Y of N” is built from; it is not the length of items.

integer format: int64

Example

{
  "items": [
    {
      "monitor_type": "http",
      "not_checked_reason": "quota",
      "group_notifications": [
        {
          "notifications": [
            {
              "type": "email",
              "details": {
                "pagerduty_severity": "critical"
              }
            }
          ]
        }
      ],
      "content_type": "json",
      "auth_type": "basic",
      "http_version": "auto"
    }
  ]
}

400

Invalid monitor_type query parameter

401

Missing or invalid authentication

500

Internal server error