Schemas

Settings of a discord channel

Settings of a microsoft_teams channel

ChannelWebhookInput

Settings of a webhook channel

ChannelPushoverInput

Settings of a pushover channel

Settings of an sms channel

Settings of a voice channel

CreateDomain

A domain to monitor.

name

string · maxLength: 255 · required

The registered domain, without www. or a path; internationalised names may be in Unicode or punycode

Example: example.com

enabled

boolean

Whether to start monitoring at once. Default true

expiry_threshold_days

integer · min: 1 · max: 365

Days before expiry, 1 to 365, that the "expires soon" alert fires. Default 30

Example: 30

alert_on_registry_changes

boolean

Raise an alert when the registrar, nameservers or DNSSEC keys change. Default true

project_id

string | null · uuid

Project to add the domain to; default the team's default project, null for none

escalation_path_ids

string[]

Escalation paths for the domain's alerts. Leave out or send [] to follow the project's paths, else the team's default path

CreateEscalationPath

A new escalation path.

name

string · maxLength: 255 · required

Name, up to 255 characters

Example: On-call engineers

description

string | null · maxLength: 1000

What the path is for, up to 1000 characters

enabled

boolean

Whether alerts follow the path. Default true

is_default

boolean

Make it the team's default path; the current default stops being one. Default false

MonitorAdvancedSettingsInput

The steps, in order, up to 99 in total. Leave out or [] for a path that does nothing yet

CreateIncident

A new incident.

title

string · maxLength: 255 · required

What's going on, up to 255 characters

Example: Checkout fails for some customers

severity

Severity · enum · required

How bad it is


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

summary

string | null

More about it, in Markdown

CreateIncidentUpdate

An update to post to an incident.

body

string · required

What's known now, in Markdown

Example: Rolled back the deploy; error rates are dropping.

status

IncidentStatus · enum

Move the incident to this status with the update; leave out to keep it


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

CreateInvite

Someone to invite to the team.

email

string · email · maxLength: 255 · required

Their email address; disposable addresses aren't accepted

Example: jane@example.com

role

TeamRole · enum

The role they get when they accept (default member); owner can't be given


`owner` Everything, including billing and deleting the team. One per team.
`manager` Everything but billing and deleting the team.
`admin` Configure monitoring, alerting, incidents and status pages; can't manage members or API tokens.
`member` Read everything; manage monitors and incidents, acknowledge alerts.
`billing` Billing and support only.

Enum values:

owner

manager

admin

member

billing

CreateMetricCondition

A new metric condition. Attach it to hosts, to projects (all their hosts), or both.

metric

MetricPoint · enum · required

The metric to watch

Enum values:

cpu_usage

system.cpu.user

system.cpu.system

system.cpu.iowait

system.cpu.stolen

system.cpu.idle

system.cpu.guest

system.cpu.interrupt

MetricCondition · required

How the metric's value is compared with threshold

threshold

number · required

Compared with the metric's value as the agent reports it, in the unit the host's charts show

Example: 90

severity

Severity · enum · required

Severity of the alert it raises


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

duration_minutes

integer · min: 0 · max: 60

Minutes, 0 to 60, the condition must hold before the alert fires. Default 0: on the first breach

Example: 5

host_ids

string[] · unique

Hosts of the team it applies to

project_ids

string[] · unique

Projects of the team whose hosts it applies to

CreateMonitor

A new monitor. Only `label`, `protocol`, `uri` and `interval` are required; everything else has a sensible default.

label

string · maxLength: 255 · required

Name shown in the dashboard, alerts and status pages

Example: Website

protocol

MonitorProtocol · enum · required

What to check; decides the format of uri


`https` An HTTPS request to the URL in `uri`, with a certificate check.
`http` An HTTP request to the URL in `uri`.
`icmp` A ping (ICMP echo) to the host in `uri`.
`tcp` A TCP connect to `host:port`, optionally with a TLS handshake and a banner keyword.
`dns` A DNS query for a record of the name in `uri`, optionally compared with expected values.
`smtp` An SMTP greeting from `host:port` (port optional), with STARTTLS by default.
`imap` An IMAP greeting from `host:port` (port optional), over TLS by default.
`pop3` A POP3 greeting from `host:port` (port optional), over TLS by default.

Enum values:

https

http

icmp

tcp

dns

smtp

imap

pop3

uri

string · maxLength: 2048 · required

The target: a URL for http and https, a host for icmp, host:port for tcp, host or host:port for mail, a DNS name for dns. Private and reserved addresses are refused. For http and https the scheme is optional: example.com/health and https://example.com/health give the same monitor, stored without the scheme. A scheme that differs from protocol (http:// for an https monitor) is refused with 422

Example: https://example.com/health

interval

MonitorInterval · enum · required

Seconds between checks. Plans set the shortest allowed interval (422 below it)

Enum values:

120

180

300

600

900

1800

enabled

boolean

Whether the monitor starts running at once. Default true

primary_location

MonitorLocation · enum

Region the checks run from. Default eu


`eu` Europe.
`us` North America.
`as` Asia.
`oc` Oceania.

Enum values:

timeout

integer · min: 1 · max: 60

Seconds before a check times out, 1 to 60. Default 30

Example: 30

max_redirects

integer | null · min: 0 · max: 10

HTTP: redirects to follow, 0 to 10; null follows up to 10

MonitorHeader[]

HTTP: up to 25 headers to send

Protocol-specific settings

allow_insecure_requests

boolean

HTTPS: accept invalid certificates for the check itself. Default false

expected_http_status_code

string | null · minLength: 1 · maxLength: 200

HTTP: status codes that count as up, comma-separated, * as a wildcard digit; null accepts 2xx and 3xx

Example: 2**,301

certificate_check_enabled

boolean

Check the TLS certificate for validity and expiry (HTTPS, and TCP or mail with TLS). Default false

certificate_expiry_threshold_days

integer · min: 1 · max: 90

Days before the certificate expires that the "expires soon" alert fires, 1 to 90. Default 14

Example: 14

performance_check_enabled

boolean

Raise alerts for slow responses. Default false

performance_hard_threshold

integer | null · min: 1 · max: 30000

Milliseconds, 1 to 30000: a response slower than this raises an alert

Example: 2000

performance_deviation_percentage

integer | null · min: 1 · max: 500

Percent, 1 to 500: alert when recent response times are this much above the earlier ones (see the two timeframes)

Example: 50

performance_deviation_timeframe

integer | null · min: 15 · max: 120

Minutes of recent checks to compare: 15, 30, 45, 60, 75, 90, 105 or 120. Must be shorter than performance_deviation_before_timeframe

Example: 30

performance_deviation_before_timeframe

integer | null · min: 60 · max: 1440

Minutes before those to compare with, a whole number of hours from 60 to 1440

Example: 240

regional_performance_enabled

boolean

Also measure response time from the regions other than primary_location, every minute (or at the interval, when longer). Shown with the monitor's performance; not used for uptime or alerts. Team plan and up. Default false

project_id

string | null · uuid

Project to add the monitor to; default the team's default project, null for none

escalation_path_ids

string[]

Escalation paths for the monitor's alerts. Leave out or send [] to follow the project's paths, else the team's default path

CreateOnCallOverride

Someone to cover a schedule for a while, at most 90 days.

member_id

string · uuid · required

The member who covers

starts_at

string · date-time · required

From when, as an ISO 8601 date and time

ends_at

string · date-time · required

Until when; after starts_at

note

string | null · maxLength: 500

Why, for the team; up to 500 characters

Example: Covering for Jane's holiday

CreateOnCallRotation

A new rotation layer of a schedule.

name

string · maxLength: 255 · required

Name, up to 255 characters

Example: Weekly primary

shift_unit

OnCallShiftUnit · enum · required

The unit of the shift length


`hours` Absolute hours: shifts keep their length across DST changes.
`days` Calendar days in the schedule's time zone: the handoff stays at the same wall-clock time.
`weeks` Calendar weeks in the schedule's time zone.

Enum values:

hours

days

weeks

shift_length

integer · min: 1 · max: 720 · required

Shift length in shift_units: up to 720 hours, 90 days or 12 weeks

Example: 1

start_date

string · date · pattern: ^\d{4}-\d{2}-\d{2}$ · required

The first handoff's date in the schedule's time zone; turns are counted from it

Example: 2026-11-02

handoff_time

string · required

Handoff time as HH:MM in the schedule's time zone

Example: 09:00

position

integer · min: 0 · max: 100

Layer order, 0 for the top. Default: below the existing layers

participant_ids

string[] · minItems: 1 · maxItems: 50 · unique

1 to 50 team members taking turns, in order

CreateOnCallSchedule

A new on-call schedule. Add rotation layers to it afterwards.

name

string · maxLength: 255 · required

Name, up to 255 characters

Example: Primary

timezone

string · required

IANA time zone the rotations' dates, handoffs and restrictions are in

Example: Europe/Amsterdam

description

string | null · maxLength: 1000

What the schedule covers, up to 1000 characters

CreateProject

A new project.

name

string · maxLength: 255 · required

Project name

Example: Production

escalation_path_ids

string[] · unique

Escalation paths for alerts of the project's monitors and hosts. Leave out for none

CreateScheduledMaintenance

Maintenance to announce on a status page.

title

string · maxLength: 255 · required

Title shown on the page, up to 255 characters

Example: Database upgrade

scheduled_at

string · date-time · required

Planned start, in the future

description

string | null

What happens and what visitors notice, in Markdown; null for none

Example: The API is read-only for about ten minutes.

scheduled_until

string | null · date-time

Planned end, after scheduled_at; null for none

automatic

boolean

true (default): starts at scheduled_at and completes at scheduled_until by itself (without a planned end, complete it by setting status). false: only changes when you set status

Components it affects; components of the status page only

notify_subscribers

boolean

Email confirmed subscribers about it (default true)

CreateStatusPage

A new status page. Settings left out get their defaults.

name

string · maxLength: 255 · required

Name, shown as the page title

Example: Acme status

slug

string · maxLength: 63 · pattern: ^[a-z0-9](?:[a-z0-9-…

The page's subdomain under spectate.page: lowercase letters, digits and dashes. Leave out to derive it from the name

Example: acme

visibility

StatusPageVisibility · enum

Who can open the page. Private pages are part of the Team plan and up (403 PLAN_FEATURE_UNAVAILABLE)


`public` Anyone.
`private` Visitors with the access link, or the password when the page has one. Part of the Team plan and up.

Enum values:

public

private

custom_domain

string | null · maxLength: 253

Your own domain for the page, such as status.acme.com; null for none. Part of paid plans (403 PLAN_FEATURE_UNAVAILABLE otherwise). A new domain needs to be verified again: see custom_domain_verification and verifyStatusPageCustomDomain

Example: status.acme.com

password_protected

boolean

Let visitors of a private page also get in with a password. Part of paid plans (403 PLAN_FEATURE_UNAVAILABLE otherwise); turning it on needs a password unless the page has one

password

string | null · minLength: 8 · maxLength: 255

The password, 8 to 255 characters. Write-only; leave out or send null to keep the current one

Example: correct-horse-battery

language

StatusPageLanguage · enum

Language of the public page and of subscriber emails


`en` English.
`nl` Dutch.
`de` German.
`fr` French.
`es` Spanish.
`it` Italian.

Enum values:

match_browser_language

boolean

Give visitors their browser's language when it is supported

StatusPageThemeInput

How the page looks and what it shows

collapsed_groups

string[] · maxItems: 100 · unique

Component groups that start collapsed; replaces the current list

automation_enabled

boolean

Monitor automation: when a monitor linked to a component goes down, open an incident "{component} is down" (major outage), and resolve it when the monitor recovers

automation_down_message

string | null · maxLength: 5000

Posted when a linked monitor goes down, up to 5000 characters; {component} and {status_page} are filled in. null for a default message

Example: We're looking into problems with {component}.

automation_recovered_message

string | null · maxLength: 5000

Posted when the monitor recovers, up to 5000 characters; null for a default message

Example: {component} is working again.

issue_reports_enabled

boolean

Let visitors report problems from the page

issue_reports_show_on_page

boolean

While reports are above the threshold, say on the page that users are reporting issues

issue_reports_alerts_enabled

boolean

Fire an alert when reports reach the threshold

issue_reports_threshold

integer · min: 1 · max: 1000

Reports within issue_reports_window_minutes that count as above the threshold, 1 to 1000

issue_reports_window_minutes

integer · min: 1 · max: 1440

The window for the threshold, in minutes, 1 to 1440

traffic_alerts_enabled

boolean

Alert when views reach traffic_alert_multiplier times the usual views for the time of day

traffic_alert_multiplier

integer · min: 2 · max: 100

How many times the usual views alert, 2 to 100

traffic_alert_min_views

integer · min: 10 · max: 1000000

Fewer views than this in 15 minutes never alert, 10 to 1000000

escalation_path_ids

string[]

Escalation paths for the page's alerts (issue reports and traffic spikes); replaces the current ones. Empty for the team's default path

CreateStatusPageComponent

A new component of a status page. Its status follows incidents and maintenance.

name

string · maxLength: 255 · required

Name shown on the page

Example: API

description

string | null

Text shown with the name; null for none

Example: Our public REST API.

group

string | null · maxLength: 255

Group to show the component in; null for the page's default group

Example: Backend

position

integer · min: 0

Order within the group, lowest first, from 0. New components go last when left out

monitor_id

string | null · uuid

A monitor of the team whose uptime history and response times the component shows, and whose downtime opens an incident when monitor automation is on; null for none

show_latency_chart

boolean

Show the monitor's response time chart under the component

CreateStatusPageImpactPeriod

A period during which a public incident affected a component. Periods of one component may not overlap within the incident.

component_id

string · uuid · required

A component of the status page

impact

ComponentImpact · enum · required

How badly


`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

degraded_performance

partial_outage

major_outage

started_at

string · date-time · required

When it started, now or in the past

ended_at

string | null · date-time

When it ended, after started_at and not in the future; null or left out while it goes on

CreateStatusPageIncident

A new public incident on a status page.

title

string · maxLength: 255 · required

Title shown on the page

Example: Elevated API error rates

status

IncidentStatus · enum · required

Where it stands


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

body

string | null

The first update, in Markdown. Leave out to create the incident without one

Example: We're looking into errors on the API.

published_at

string | null · date-time

When it happened, now or in the past (default now)

impact

ImpactLevel · enum

The incident's own impact, used while it affects no components; otherwise it follows the components


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

incident_id

string | null · uuid

The team's incident this one is for; null for none. Publishing from that incident (publishIncidentStatusPageUpdate) then posts here

The components it affects as of published_at, with their impact; components left out are no longer affected. Leave out to keep the impacts as they are; send [] for none. Components of the status page only

notify_subscribers

boolean

Email confirmed subscribers about the new incident (default true). false changes the page without emails

CreateStatusPageIncidentUpdate

An update to publish on a public incident.

status

IncidentStatus · enum · required

The incident's status with this update. resolved closes every open impact period


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

body

string · required

The message, in Markdown

Example: We found the cause and are rolling out a fix.

published_at

string | null · date-time

When it happened, now or in the past (default now)

notify_subscribers

boolean

Email confirmed subscribers about the update (default true). false changes the page without emails

DnsMatch

string · enum

Enum values:

contains

exact

How a DNS monitor compares the answer with dns_expected_values.


`contains` Every expected value is in the answer; the answer may have more.
`exact` The answer is exactly the expected values, in any order.

DnsRecordType

string · enum

Enum values:

AAAA

CNAME

TXT

CAA

SOA

The DNS record type a DNS monitor queries.

Domain

A monitored domain registration: Spectate looks it up at its registry (RDAP or WHOIS) every few hours and alerts before it expires and, optionally, when its registrar, nameservers or DNSSEC change.

id

string · uuid · required

Domain ID

name

string · required

The registered domain, lowercase, internationalised names in punycode

Example: example.com

enabled

boolean · required

Whether the domain is monitored. false when paused by a member or by the system (see paused_reason)

required

registrar

string | null · required

The registrar, as the registry reports it

Example: Example Registrar, Inc.

registered_at

string | null · date-time · required

When the domain was registered

expires_at

string | null · date-time · required

When the registration expires

registry_changed_at

string | null · date-time · required

When the registration last changed at the registry

days_until_expiration

integer | null · required

Days until the registration expires; negative once expired

Example: 212

is_expired

boolean · required

Whether the registration has expired

is_expiring_soon

boolean · required

Whether it expires within expiry_threshold_days

nameservers

array | null · required

The domain's nameservers

dnssec

boolean · required

Whether the registry publishes DNSSEC keys for the domain

array | null · required

The DNSSEC delegation signer records

statuses

array | null · required

EPP status codes, for example client transfer prohibited

required

last_checked_at

string | null · date-time · required

When the registry was last checked

expiry_threshold_days

integer · required

Days before expiry that the "expires soon" alert fires

Example: 30

alert_on_registry_changes

boolean · required

Whether changes of the registrar, nameservers or DNSSEC keys raise an alert

project_id

string | null · uuid · required

The project the domain belongs to; null for none

EscalationPathReference[] · required

Escalation paths the domain's alerts follow. Empty: the project's paths, else the team's default path

created_at

string · date-time · required

When monitoring of the domain started

DomainDnssecKey

A DNSSEC delegation signer (DS) record the registry publishes for the domain.

key_tag

integer | null · required

Key tag of the DNSKEY the record points at

Example: 2371

algorithm

integer | null · required

DNSSEC algorithm number, for example 13 (ECDSA P-256 with SHA-256)

Example: 13

digest_type

integer | null · required

Digest type number, for example 2 (SHA-256)

Example: 2

digest

string | null · required

The digest, in hexadecimal

DomainLookupError

string · enum

Enum values:

unsupported

not_registered

unavailable

Why the last registry lookup of a domain gave no data.


`unsupported` The registry publishes neither RDAP nor WHOIS (e.g. .gr, .vn).
`not_registered` The registry has no such domain: not registered (or a subdomain was entered).
`unavailable` The registry didn't answer usefully, even after retries; tried again at the next check.

DomainRegistrySource

string · enum

Enum values:

rdap

whois

Where a domain's registry data came from.


`rdap` The registry's RDAP service.
`whois` The registry's WHOIS server, for registries without RDAP.

DomainView

string · enum

Enum values:

attention

unsigned

paused

Domain list views; a domain can be in several.


`attention` Expired, or expiring within its `expiry_threshold_days`.
`unsigned` No DNSSEC keys.
`paused` Monitoring is paused.

EscalationCondition

A condition of a branch. The branch continues on its `then` side when every condition holds, else on its `else` side.

field

EscalationConditionField · enum · required

What is tested


`severity`
`working_hours` Whether the branch is reached inside the path's working hours.
`source` The kind of check or integration that raised the alert (uptime, certificate, external…).
`label` A label sent along by an alert source (`metadata.labels`), e.g. `env=production`.

Enum values:

severity

working_hours

source

label

operator

EscalationConditionOperator · enum · required

How it's tested; each field allows some operators (see the operator values)


`is_one_of` The value is one of `values`.
`is_not_one_of` The value is none of `values`.
`within` Reached inside the path's working hours.
`outside` Reached outside the path's working hours.
`exists` The label is set.
`not_exists` The label isn't set.

Enum values:

is_one_of

is_not_one_of

within

outside

exists

not_exists

key

string | null · required

label conditions: the label name sent by the alert source

Example: env

values

string[] · required

Values to match, for is_one_of and is_not_one_of: severities (critical, high, medium, low), sources (uptime, performance, certificate, dns, domain, host_metric, status_page, external, or external:<tool> such as external:grafana) or label values

description

string · required

The condition in words

Example: Severity is Critical or High

EscalationConditionField

string · enum

Enum values:

severity

working_hours

source

label

What an escalation branch can test about an alert (incident.io-style "if/else" branches).


`severity`
`working_hours` Whether the branch is reached inside the path's working hours.
`source` The kind of check or integration that raised the alert (uptime, certificate, external…).
`label` A label sent along by an alert source (`metadata.labels`), e.g. `env=production`.

EscalationConditionInput

A condition of a branch. `severity` and `source` take `is_one_of` and `is_not_one_of` with `values`; `working_hours` takes `within` and `outside`; `label` takes a `key` and any of the four list and presence operators.

field

EscalationConditionField · enum · required

What to test


`severity`
`working_hours` Whether the branch is reached inside the path's working hours.
`source` The kind of check or integration that raised the alert (uptime, certificate, external…).
`label` A label sent along by an alert source (`metadata.labels`), e.g. `env=production`.

Enum values:

severity

working_hours

source

label

operator

EscalationConditionOperator · enum · required

How to test it


`is_one_of` The value is one of `values`.
`is_not_one_of` The value is none of `values`.
`within` Reached inside the path's working hours.
`outside` Reached outside the path's working hours.
`exists` The label is set.
`not_exists` The label isn't set.

Enum values:

is_one_of

is_not_one_of

within

outside

exists

not_exists

key

string | null · maxLength: 100

label conditions: the label name sent by the alert source, up to 100 characters

Example: env

values

string[] · maxItems: 20

Up to 20 values to match, for is_one_of and is_not_one_of: severities, sources (uptime, external:grafana…) or label values

EscalationConditionInputRequest

field

EscalationConditionField · enum · required

What to test


`severity`
`working_hours` Whether the branch is reached inside the path's working hours.
`source` The kind of check or integration that raised the alert (uptime, certificate, external…).
`label` A label sent along by an alert source (`metadata.labels`), e.g. `env=production`.

Enum values:

severity

working_hours

source

label

operator

EscalationConditionOperator · enum · required

How to test it


`is_one_of` The value is one of `values`.
`is_not_one_of` The value is none of `values`.
`within` Reached inside the path's working hours.
`outside` Reached outside the path's working hours.
`exists` The label is set.
`not_exists` The label isn't set.

Enum values:

is_one_of

is_not_one_of

within

outside

exists

not_exists

key

string | null · maxLength: 100

label conditions: the label name sent by the alert source, up to 100 characters

Example: env

values

string[] · maxItems: 20

Up to 20 values to match, for is_one_of and is_not_one_of: severities, sources (uptime, external:grafana…) or label values

EscalationConditionOperator

string · enum

Enum values:

is_one_of

is_not_one_of

within

outside

exists

not_exists

How an escalation branch condition tests its field. severity and source take is_one_of and is_not_one_of; working_hours takes within and outside; label takes all four list and presence operators.


`is_one_of` The value is one of `values`.
`is_not_one_of` The value is none of `values`.
`within` Reached inside the path's working hours.
`outside` Reached outside the path's working hours.
`exists` The label is set.
`not_exists` The label isn't set.

EscalationDelayUntil

string · enum

Enum values:

working_hours_start

working_hours_end

Instead of a fixed delay, an escalation step can wait for the path's working hours to begin (no wait during working hours) or to end (no wait outside them).


`working_hours_start` Wait until working hours begin; no wait during working hours.
`working_hours_end` Wait until working hours end; no wait outside them.

EscalationPath

An escalation path: what happens when an alert fires, step by step, until someone acknowledges it. Monitors, hosts, domains and alert sources use their own paths, their project's, or the team's default path.

id

string · uuid · required

Escalation path ID

name

string · required

Name

Example: On-call engineers

description

string | null · required

What the path is for

enabled

boolean · required

Whether alerts follow the path. Alerts of a disabled path don't escalate

is_default

boolean · required

Whether this is the team's default path, for resources without paths of their own or of their project. One per team

WorkingHours · required

The working hours that branch conditions and "until working hours" waits use

EscalationStep[] · required

The steps, in order; a branch continues in its then and else steps

EscalationPathWarning[] · required

Problems that keep the path from reaching someone; empty when it can deliver

created_at

string · date-time · required

When the path was created

EscalationPathReference

An escalation path a resource's alerts follow.

id

string · uuid · required

Escalation path ID

name

string · required

Escalation path name

Example: On-call engineers

EscalationPathWarning

A delivery problem of an escalation path.

code

EscalationPathWarningCode · enum · required

What's wrong


`no_deliverable_steps` Every notify step targets a disabled channel or a deleted schedule: alerts reach nobody.
`disabled_channels` Some channels are disabled; their steps are skipped.
`missing_schedules` An on-call step's schedule was deleted; the step is skipped.

Enum values:

no_deliverable_steps

disabled_channels

missing_schedules

message

string · required

The problem in words

Example: Channel Ops Slack is disabled and will be skipped.

EscalationPathWarningCode

string · enum

Enum values:

no_deliverable_steps

disabled_channels

missing_schedules

A problem that keeps an escalation path from reaching someone.


`no_deliverable_steps` Every notify step targets a disabled channel or a deleted schedule: alerts reach nobody.
`disabled_channels` Some channels are disabled; their steps are skipped.
`missing_schedules` An on-call step's schedule was deleted; the step is skipped.

EscalationStep

A step of an escalation path. Steps run in order, each after its wait. A `branch` is always the last step of its list: alerts go on with its `then` steps when its conditions hold, else with its `else` steps.

type

EscalationStepAction · enum · required

What the step does


`notify_channel` Send the alert to a channel.
`create_incident` Declare an incident for the alert.
`notify_on_call` Page whoever is on call in a schedule, by email and their own phone.
`branch` If/else: alerts meeting every condition continue on its `then` side, the rest on its `else` side.

Enum values:

notify_channel

create_incident

notify_on_call

branch

delay_minutes

integer | null · required

Minutes to wait before the step runs; null for a branch (decided at once) or when delay_until applies

Example: 5

required

array | null · required

branch steps: the conditions; all must hold for then

array | null · required

branch steps: what happens when the conditions hold

array | null · required

branch steps: what happens otherwise

EscalationStepAction

string · enum

Enum values:

notify_channel

create_incident

notify_on_call

branch

What an escalation step does.


`notify_channel` Send the alert to a channel.
`create_incident` Declare an incident for the alert.
`notify_on_call` Page whoever is on call in a schedule, by email and their own phone.
`branch` If/else: alerts meeting every condition continue on its `then` side, the rest on its `else` side.

EscalationStepInput

A step to run. `notify_channel` needs `channel_id`, `notify_on_call` needs `schedule_id`, `branch` needs `conditions` and ends its list: what follows goes in its `then` and `else`.

type

EscalationStepAction · enum · required

What the step does


`notify_channel` Send the alert to a channel.
`create_incident` Declare an incident for the alert.
`notify_on_call` Page whoever is on call in a schedule, by email and their own phone.
`branch` If/else: alerts meeting every condition continue on its `then` side, the rest on its `else` side.

Enum values:

notify_channel

create_incident

notify_on_call

branch

delay_minutes

integer · min: 0 · max: 1440

Minutes, 0 to 1440, to wait before the step runs. Default 0. Not for branches

Example: 5

channel_id

string · uuid

notify_channel: the channel to send to

schedule_id

string · uuid

notify_on_call: the schedule whose person on call is paged (on-call is part of the Team plan and up)

EscalationConditionInput[]

branch: 1 to 10 conditions; all must hold for then

branch: the steps when the conditions hold; empty ends the escalation there

branch: the steps otherwise; empty ends the escalation there

EscalationStepInputRequest

A step to run. `notify_channel` needs `channel_id`, `notify_on_call` needs `schedule_id`, `branch` needs `conditions` and ends its list: what follows goes in its `then` and `else`.

type

EscalationStepAction · enum · required

What the step does


`notify_channel` Send the alert to a channel.
`create_incident` Declare an incident for the alert.
`notify_on_call` Page whoever is on call in a schedule, by email and their own phone.
`branch` If/else: alerts meeting every condition continue on its `then` side, the rest on its `else` side.

Enum values:

notify_channel

create_incident

notify_on_call

branch

delay_minutes

integer · min: 0 · max: 1440

Minutes, 0 to 1440, to wait before the step runs. Default 0. Not for branches

Example: 5

channel_id

string · uuid

notify_channel: the channel to send to

schedule_id

string · uuid

notify_on_call: the schedule whose person on call is paged (on-call is part of the Team plan and up)

EscalationConditionInput[]

branch: 1 to 10 conditions; all must hold for then

branch: the steps when the conditions hold; empty ends the escalation there

branch: the steps otherwise; empty ends the escalation there

HealthStatus

string · enum

Enum values:

warning

critical

unknown

Verdict of a host health check.


`ok` Nothing to do.
`warning` Needs attention soon: a reboot or security updates pending, the clock not synchronised.
`critical` Needs attention now: a service failed, processes were killed for lack of memory.
`unknown` The agent does not report it (yet): an older agent, or nothing to report on this host.

Host

A server with the Spectate agent, which reports CPU, memory, disk and network metrics. Hosts are added by installing the agent with an install token, not through the API.

id

string · uuid · required

Host ID

label

string | null · required

Name shown in the dashboard and alerts; the agent sets the hostname until you change it

Example: web-1

state

HostState · enum · required

Whether the agent reports


`up` Reported in the last 15 minutes.
`stale` Reported before, but not in the last 15 minutes.
`pending` The agent never reported.

Enum values:

stale

pending

object | null · required

What the agent reports about the system, as key-value pairs (for example hostname, os, platform, kernel_version); the keys depend on the agent version and platform

cpu_info

array | null · required

The CPUs, each as key-value pairs (for example model_name, cores, mhz)

agent_version

string | null · required

Version of the agent

Example: 2.0.3

agent_update_version

string | null · required

A newer agent release, when the host runs an older one (sudo spectate-agent update)

clock_offset_ms

integer | null · required

Server time minus the host's clock, in milliseconds, as last measured by a v2 agent; null when unknown

last_seen_at

string | null · date-time · required

When the agent last reported

no_data_alert_minutes

integer | null · required

Minutes without data before the "no data received" alert fires; null when that alert is off

Example: 10

required

project_id

string | null · uuid · required

The project the host belongs to; null for none

EscalationPathReference[] · required

Escalation paths the host's alerts follow. Empty: the project's paths, else the team's default path

created_at

string · date-time · required

When the host registered

HostHealthCheckResult[] · required

Health checks (agent 2.1 or later): failed services, out of memory, reboot, updates, clock, file handles and pressure. Checks of collectors that are switched off, and Linux-only checks on other systems, are left out

health_issues

integer · required

Number of health checks that need attention (warning or critical)

HostHealthCheck

string · enum

Enum values:

services

oom

reboot

updates

clock

files

pressure

A host health check, computed from what agent 2.1+ reports.


`services` Failed systemd services.
`oom` Processes killed for lack of memory (OOM kills) in the last 24 hours.
`reboot` The system needs a reboot (after kernel or library updates).
`updates` Pending package updates, security updates in particular.
`clock` Whether the clock is synchronised (NTP) and close to the server's.
`files` Open file handles against the system's maximum.
`pressure` Tasks stalled waiting for memory or I/O (pressure stall information).

HostHealthCheckResult

One health check of a host, computed from what agent 2.1 or later reports.

check

HostHealthCheck · enum · required

What is checked


`services` Failed systemd services.
`oom` Processes killed for lack of memory (OOM kills) in the last 24 hours.
`reboot` The system needs a reboot (after kernel or library updates).
`updates` Pending package updates, security updates in particular.
`clock` Whether the clock is synchronised (NTP) and close to the server's.
`files` Open file handles against the system's maximum.
`pressure` Tasks stalled waiting for memory or I/O (pressure stall information).

Enum values:

services

oom

reboot

updates

clock

files

pressure

status

HealthStatus · enum · required

The verdict


`ok` Nothing to do.
`warning` Needs attention soon: a reboot or security updates pending, the clock not synchronised.
`critical` Needs attention now: a service failed, processes were killed for lack of memory.
`unknown` The agent does not report it (yet): an older agent, or nothing to report on this host.

Enum values:

warning

critical

unknown

detail

string | null · required

What was found, in one line

Example: nginx.service failed

since

string | null · date-time · required

Since when the problem has lasted (for out of memory: the last kill); null when there is none

HostInstallToken

An install token: agents use it to register hosts with the team. Any number of hosts can use one token until it expires or is revoked.

id

string · uuid · required

Install token ID

token

string · required

The token to give the agent's installer (--token)

Example: hct_0b7c…

expires_at

string | null · date-time · required

When the token stops registering new hosts. Hosts that registered keep reporting

created_at

string · date-time · required

When the token was created

HostReference

A host something applies to.

id

string · uuid · required

Host ID

label

string | null · required

Host name shown in the dashboard; null until set or reported

Example: web-1

HostState

string · enum

Enum values:

stale

pending

Whether a host's agent reports.


`up` Reported in the last 15 minutes.
`stale` Reported before, but not in the last 15 minutes.
`pending` The agent never reported.

ImpactLevel

string · enum

Enum values:

none

degraded_performance

partial_outage

major_outage

An impact on a status page, including none: how an incident or maintenance affects a component, or an incident's overall impact.


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Incident

An incident: a problem that affects customers, worked on by the team, with updates, roles and eventually a postmortem. Incidents are declared by hand, through the API, from Slack, or by an escalation path's `create_incident` step.

id

string · uuid · required

Incident ID

number

integer | null · required

The team's sequence number: INC-42 is 42. null for some older incidents

Example: 42

reference

string · required

The human reference

Example: INC-42

title

string · required

What's going on

Example: Checkout fails for some customers

summary

string | null · required

More about it, in Markdown

severity

Severity · enum · required

How bad it is


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

status

IncidentStatus · enum · required

Where it stands


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

started_at

string · date-time · required

When it started

acknowledged_at

string | null · date-time · required

When someone acknowledged it

resolved_at

string | null · date-time · required

When it was resolved

alert_id

string | null · uuid · required

The alert that opened it, for incidents declared by an escalation path

IncidentRoleAssignment[] · required

Who has which role

created_at

string · date-time · required

When it was declared

IncidentPostmortem

An incident's postmortem: what happened, why, and what changes.

content

string | null · required

The postmortem, in Markdown

published

boolean · required

Whether it is marked as published (final). It is not shown on status pages

updated_at

string · date-time · required

When it last changed

IncidentPostmortemResult

An incident's postmortem, if it has one.

required

IncidentRole

string · enum

Enum values:

commander

communications_lead

A role in handling an incident; one member each.


`commander` Leads the response and makes the calls.
`communications_lead` Keeps customers and the company informed, for example through the status page.

IncidentRoleAssignment

Who has a role in an incident.

role

IncidentRole · enum · required

The role


`commander` Leads the response and makes the calls.
`communications_lead` Keeps customers and the company informed, for example through the status page.

Enum values:

commander

communications_lead

MemberReference · required

The member who has it

IncidentStatus

string · enum

Enum values:

investigating

identified

monitoring

resolved

Where an incident stands.


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

IncidentTimelineEntry

Something that happened in an incident.

description

string · required

What happened

Example: Incident acknowledged

actor

string | null · required

Who did it: a member's name, API token "…", or Spectate staff; null for Spectate itself

Example: Jane Doe

required

created_at

string · date-time · required

When it happened

IncidentUpdate

An update posted to an incident: what's known now, and its status at that moment.

id

string · uuid · required

Update ID

status

IncidentStatus · enum · required

The incident's status with this update


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

body

string · required

The update, in Markdown

Example: Rolled back the deploy; error rates are dropping.

created_at

string · date-time · required

When it was posted

Invite

An invite to join the team, sent by email. The invitee accepts it in the dashboard with an account for that address.

id

string · uuid · required

Invite ID

email

string · email · required

The invitee's email address

Example: jane@example.com

role

TeamRole · enum · required

The role they get when they accept


`owner` Everything, including billing and deleting the team. One per team.
`manager` Everything but billing and deleting the team.
`admin` Configure monitoring, alerting, incidents and status pages; can't manage members or API tokens.
`member` Read everything; manage monitors and incidents, acknowledge alerts.
`billing` Billing and support only.

Enum values:

owner

manager

admin

member

billing

expired

boolean · required

Whether it expired without being accepted. Expired invites don't count towards the member limit and can be sent again

expires_at

string · date-time · required

When it expires, 7 days after it was sent

created_at

string · date-time · required

When it was sent

Member

A member of the team. Use member IDs for on-call rotations and incident roles.

id

string · uuid · required

Member (user) ID

name

string · required

Name

Example: Jane Doe

email

string · email · required

Email address

Example: jane@example.com

role

TeamRole · enum · required

Role in the team


`owner` Everything, including billing and deleting the team. One per team.
`manager` Everything but billing and deleting the team.
`admin` Configure monitoring, alerting, incidents and status pages; can't manage members or API tokens.
`member` Read everything; manage monitors and incidents, acknowledge alerts.
`billing` Billing and support only.

Enum values:

owner

manager

admin

member

billing

joined_at

string | null · date-time · required

When they joined the team

MemberReference

A member of the team.

id

string · uuid · required

Member (user) ID

name

string · required

Name

Example: Jane Doe

email

string · email · required

Email address

Example: jane@example.com

MetricCondition

An alert rule for host metrics, for example "CPU usage above 90% for 5 minutes". It applies to its hosts and to every host of its projects.

id

string · uuid · required

Metric condition ID

metric

MetricPoint · enum · required

The metric to watch

Enum values:

cpu_usage

system.cpu.user

system.cpu.system

system.cpu.iowait

system.cpu.stolen

system.cpu.idle

system.cpu.guest

system.cpu.interrupt

MetricCondition · required

How the metric's value is compared with threshold


`greater_than` Fires when the value is above the threshold.
`less_than` Fires when the value is below the threshold.
`equals` Fires when the value equals the threshold.

threshold

number · required

Compared with the metric's value as the agent reports it, in the unit the host's charts show (percent for cpu_usage, memory_usage and disk_usage)

Example: 90

duration_minutes

integer · required

Minutes the condition must hold before the alert fires; 0 fires on the first breach

Example: 5

severity

Severity · enum · required

Severity of the alert it raises


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

HostReference[] · required

Hosts it applies to

ProjectReference[] · required

Projects whose hosts it applies to

MetricPoint

string · enum

Enum values:

cpu_usage

system.cpu.user

system.cpu.system

system.cpu.iowait

system.cpu.stolen

system.cpu.idle

system.cpu.guest

system.cpu.interrupt

A host metric the agent reports. cpu_usage, memory_usage, disk_usage, load_average, network_in and network_out are the summaries the dashboard charts; the system.* metrics are the agent's raw values.

Monitor

A monitor: a check of a URL, host or DNS name from Spectate's regions, with its settings and latest results.

id

string · uuid · required

Monitor ID

label

string · required

Name shown in the dashboard, alerts and status pages

Example: Website

protocol

MonitorProtocol · enum · required

What is checked; decides the format of uri


`https` An HTTPS request to the URL in `uri`, with a certificate check.
`http` An HTTP request to the URL in `uri`.
`icmp` A ping (ICMP echo) to the host in `uri`.
`tcp` A TCP connect to `host:port`, optionally with a TLS handshake and a banner keyword.
`dns` A DNS query for a record of the name in `uri`, optionally compared with expected values.
`smtp` An SMTP greeting from `host:port` (port optional), with STARTTLS by default.
`imap` An IMAP greeting from `host:port` (port optional), over TLS by default.
`pop3` A POP3 greeting from `host:port` (port optional), over TLS by default.

Enum values:

https

http

icmp

tcp

dns

smtp

imap

pop3

uri

string · required

The target as stored: a URL without the scheme for http and https (the protocol is the scheme), a host for icmp, host:port for tcp and mail, a DNS name for dns. url has the full URL

Example: example.com/health

url

string · required

The target as the dashboard shows it, for example smtp://mail.example.com:587 for SMTP

Example: https://example.com/health

interval

MonitorInterval · enum · required

Seconds between checks

Enum values:

120

180

300

600

900

1800

primary_location

MonitorLocation · enum · required

Region the checks run from; a failed check is repeated from the same region about 5 seconds later before the monitor goes down


`eu` Europe.
`us` North America.
`as` Asia.
`oc` Oceania.

Enum values:

timeout

integer · required

Seconds before a check times out

Example: 30

max_redirects

integer | null · required

HTTP: redirects followed before the check fails; null follows up to 10

MonitorHeader[] · required

HTTP: headers sent with every check

MonitorAdvancedSettings · required

Protocol-specific settings

enabled

boolean · required

Whether the monitor runs. false when paused by a member or by the system (see paused_reason)

required

allow_insecure_requests

boolean · required

HTTPS: whether invalid certificates are accepted for the check itself (the certificate check still reports them)

expected_http_status_code

string | null · required

HTTP: status codes that count as up, comma-separated, * as a wildcard digit (2**,301); null accepts 2xx and 3xx

Example: 2**,301

certificate_check_enabled

boolean · required

Whether the TLS certificate is checked for validity and expiry

certificate_expiry_threshold_days

integer · required

Days before the certificate expires that the "expires soon" alert fires

Example: 14

uses_tls

boolean · required

Whether a check reports a certificate: HTTPS, and TCP or mail with TLS

required

certificate_expires_at

string | null · date-time · required

When the certificate expires

certificate_valid_from

string | null · date-time · required

When the certificate became valid

certificate_issuer

string | null · required

Who issued the certificate

Example: R11

certificate_error

string | null · required

Why the certificate isn't valid, with certificate_status invalid or expired

certificate_checked_at

string | null · date-time · required

When the certificate was last checked

performance_check_enabled

boolean · required

Whether slow responses raise alerts (performance_hard_threshold, performance_deviation_percentage)

performance_hard_threshold

integer | null · required

Milliseconds: a response slower than this raises an alert

Example: 2000

performance_deviation_percentage

integer | null · required

Percent: an alert fires when the average response time of the last performance_deviation_timeframe minutes is this much above the average of the performance_deviation_before_timeframe minutes before

Example: 50

performance_deviation_timeframe

integer | null · required

Minutes of recent checks compared (15 to 120, in steps of 15)

Example: 30

performance_deviation_before_timeframe

integer | null · required

Minutes before those that they're compared with (60 to 1440, in whole hours)

Example: 240

regional_performance_enabled

boolean · required

uptime_status

UptimeStatus · enum · required

Whether the target answers, from the latest checks


`pending` Not checked yet.
`up` The latest check succeeded.
`down` Checks fail: a failed check, repeated from the same region about 5 seconds later, failed again.

Enum values:

pending

down

uptime_status_changed_at

string | null · date-time · required

When uptime_status last changed

last_checked_at

string | null · date-time · required

When the monitor was last checked

response_time

integer | null · required

Milliseconds the latest check took

Example: 182

required

MonitorUptime · required

Uptime over rolling windows

project_id

string | null · uuid · required

The project the monitor belongs to; null for none

EscalationPathReference[] · required

Escalation paths the monitor's alerts follow. Empty: the project's paths, else the team's default path

created_at

string · date-time · required

When the monitor was created

updated_at

string · date-time · required

When the monitor's settings last changed

MonitorAdvancedSettings

Protocol-specific settings of a monitor. Settings that don't apply to its protocol are `null`. The password is never returned; `has_auth_password` says whether one is set.

connection_timeout

integer | null · required

Seconds to wait for the connection; null uses the monitor's timeout

required

auth_user

string | null · required

HTTP: user name for basic and digest

has_auth_password

boolean · required

HTTP: whether a password or bearer token is stored

keyword

string | null · required

HTTP: text the response body must contain (or must not, with keyword_inverted). TCP: the same for the server's banner

Example: OK

keyword_inverted

boolean | null · required

Whether the check fails when keyword is found, instead of when it's missing

useragent

string | null · required

HTTP: the User-Agent header; null sends Spectate's

required

dns_resolver

string | null · required

DNS: the resolver to query, as ip or ip:port; null uses the checker's resolver

Example: 1.1.1.1

dns_authoritative

boolean | null · required

DNS: whether to ask the zone's authoritative nameservers directly

dns_expected_values

array | null · required

DNS: values the answer must contain or equal (see dns_match); null or empty accepts any answer

required

dns_alert_on_change

boolean | null · required

DNS: whether a changed answer raises an alert

MonitorAdvancedSettingsInput

Protocol-specific settings to set. On update, settings you leave out keep their value; send `null` to reset one to its default.

connection_timeout

integer | null · min: 1 · max: 60

Seconds to wait for the connection, 1 to 60; null uses the monitor's timeout

auth_user

string | null · maxLength: 255

HTTP: user name for basic and digest

auth_password

string | null · maxLength: 2500

HTTP: password, or the token for bearer. Write-only: never returned. Leave out or send an empty string to keep the stored one

keyword

string | null · maxLength: 255

HTTP: text the response body must contain (or must not, with keyword_inverted). TCP: the same for the server's banner

Example: OK

keyword_inverted

boolean | null

Fail the check when keyword is found, instead of when it's missing

useragent

string | null · maxLength: 255

HTTP: the User-Agent header; null sends Spectate's

dns_resolver

string | null · maxLength: 64

DNS: the resolver to query, as a public ip or ip:port; null uses the checker's resolver

Example: 1.1.1.1

dns_authoritative

boolean | null

DNS: ask the zone's authoritative nameservers directly

dns_expected_values

array | null · maxItems: 20

DNS: up to 20 values the answer must contain or equal (see dns_match)

dns_alert_on_change

boolean | null

DNS: raise an alert when the answer changes

MonitorAuthMethod

string · enum

Enum values:

basic

bearer

digest

How an HTTP monitor authenticates its request.


`basic` HTTP Basic: `auth_user` and `auth_password`.
`bearer` `Authorization: Bearer <auth_password>`.
`digest` HTTP Digest: `auth_user` and `auth_password`.

MonitorCheckDetails

Protocol details of the latest check, as the checker reported them. Fields that don't apply to the protocol are `null`.

banner

string | null · required

TCP and mail: the first line of the server's greeting or banner

Example: 220 mail.example.com ESMTP

reply_code

integer | null · required

SMTP: the greeting's reply code

Example: 220

starttls

boolean | null · required

Mail: whether the session was upgraded with STARTTLS (or STLS)

tls_version

string | null · required

The negotiated TLS version

Example: TLS 1.3

cipher_suite

string | null · required

The negotiated cipher suite

Example: TLS_AES_128_GCM_SHA256

dns_answers

array | null · required

DNS: the answer's values of the queried record type, sorted

dns_server

string | null · required

DNS: the server that answered, as ip:port

Example: 1.1.1.1:53

dns_rcode

string | null · required

DNS: the response code

Example: NOERROR

MonitorHeader

A request header an HTTP monitor sends.

key

string · required

Header name

Example: X-Health-Check

value

string · required

Header value

Example: spectate

MonitorHttpMethod

string · enum

Enum values:

HEAD

GET

POST

PUT

PATCH

DELETE

The HTTP method an HTTP monitor sends.

MonitorInterval

integer · enum

Enum values:

120

180

300

600

900

1800

Seconds between checks. Plans set the shortest interval a team may use.

MonitorLocation

string · enum

Enum values:

A region checks run from. A monitor's primary region decides its uptime and alerts (a failed check is repeated from the same region about 5 seconds later before it is marked down); on the Team plan and up the other regions can also measure its response time.


`eu` Europe.
`us` North America.
`as` Asia.
`oc` Oceania.

MonitorProtocol

string · enum

Enum values:

https

http

icmp

tcp

dns

smtp

imap

pop3

What a monitor checks, and so the format of its uri.


`https` An HTTPS request to the URL in `uri`, with a certificate check.
`http` An HTTP request to the URL in `uri`.
`icmp` A ping (ICMP echo) to the host in `uri`.
`tcp` A TCP connect to `host:port`, optionally with a TLS handshake and a banner keyword.
`dns` A DNS query for a record of the name in `uri`, optionally compared with expected values.
`smtp` An SMTP greeting from `host:port` (port optional), with STARTTLS by default.
`imap` An IMAP greeting from `host:port` (port optional), over TLS by default.
`pop3` A POP3 greeting from `host:port` (port optional), over TLS by default.

MonitorReference

A monitor something follows.

id

string · uuid · required

Monitor ID

label

string · required

Monitor name

Example: Website

MonitorState

string · enum

Enum values:

down

pending

paused

A monitor's state as the dashboard shows it: paused when disabled, otherwise its uptime status.


`down` Checks fail: a failed check, repeated from the same region about 5 seconds later, failed again.
`pending` Not checked yet.
`up` The latest check succeeded.
`paused` Disabled by a member or the system.

MonitorUptime

Share of successful checks, in percent, over rolling windows; `null` until there is data.

twenty_four_hours

number | null · required

Last 24 hours

Example: 99.95

seven_days

number | null · required

Last 7 days

thirty_days

number | null · required

Last 30 days

ninety_days

number | null · required

Last 90 days

NotificationCredits

Credits for SMS and voice notifications. The monthly allowance is spent first; bought credits never expire.

balance

integer · required

Credits that can be spent now: what's left of this month's allowance plus bought credits

allowance

integer · required

What's left of this month's allowance

purchased

integer · required

Bought credits left

monthly

integer · required

The plan's monthly allowance; 0 during the trial

reset_at

string | null · date-time · required

When the allowance was last reset; null before the first reset

OnCallNow

Who is on call at a moment.

required

OnCallOverride

Someone covering a schedule for a while, whatever the rotations say.

id

string · uuid · required

Override ID

MemberReference · required

Who covers

starts_at

string · date-time · required

From when

ends_at

string · date-time · required

Until when

note

string | null · required

Why, for the team

Example: Covering for Jane's holiday

required

created_at

string · date-time · required

When it was added

OnCallRestrictions

When a rotation layer covers, each week; outside it the layers below take over.

days

integer[] · required

ISO weekdays the layer covers, 1 (Monday) to 7 (Sunday)

start_time

string | null · required

From when on those days, as HH:MM in the schedule's time zone; null with end_time null for the whole day

Example: 09:00

end_time

string | null · required

Until when, as HH:MM; before start_time runs past midnight

Example: 17:00

OnCallRestrictionsInput

When a rotation layer covers, each week.

days

integer[] · minItems: 1 · maxItems: 7 · unique · required

ISO weekdays the layer covers, 1 (Monday) to 7 (Sunday), at least one

start_time

string | null

From when on those days, as HH:MM in the schedule's time zone; leave out with end_time for the whole day

Example: 09:00

end_time

string | null

Until when, as HH:MM

Example: 17:00

OnCallRotation

A rotation layer: its participants take turns of `shift_length` `shift_unit`s, from `start_date` at `handoff_time`. Where layers overlap, the top layer (`position` 0) wins.

id

string · uuid · required

Rotation ID

name

string · required

Name

Example: Weekly primary

position

integer · required

Layer order; 0 is the top layer and wins where layers overlap

shift_unit

OnCallShiftUnit · enum · required

The unit of the shift length


`hours` Absolute hours: shifts keep their length across DST changes.
`days` Calendar days in the schedule's time zone: the handoff stays at the same wall-clock time.
`weeks` Calendar weeks in the schedule's time zone.

Enum values:

hours

days

weeks

shift_length

integer · required

Shift length in shift_units

Example: 1

start_date

string · date · required

The first handoff's date in the schedule's time zone

Example: 2026-11-02

handoff_time

string · required

Handoff time as HH:MM in the schedule's time zone

Example: 09:00

required

MemberReference[] · required

The members taking turns, in order

created_at

string · date-time · required

When the rotation was created

OnCallSchedule

An on-call schedule: rotation layers and overrides that decide who gets paged when an escalation path's `notify_on_call` step runs.

id

string · uuid · required

Schedule ID

name

string · required

Name

Example: Primary

description

string | null · required

What the schedule covers

timezone

string · required

IANA time zone the rotations' dates, handoffs and restrictions are in

Example: Europe/Amsterdam

required

next_handoff_at

string | null · date-time · required

When the person on call changes next, within 4 weeks

required

OnCallRotation[] · required

Rotation layers, top first

OnCallOverride[] · required

Overrides that haven't ended

EscalationPathReference[] · required

Escalation paths with a step that pages this schedule

created_at

string · date-time · required

When the schedule was created

OnCallScheduleReference

An on-call schedule something pages.

id

string · uuid · required

On-call schedule ID

name

string · required

Schedule name

Example: Primary

OnCallShift

A stretch of time someone is on call.

required

starts_at

string · date-time · required

When the shift starts

ends_at

string · date-time · required

When it ends

source

OnCallShiftSource · enum · required

Whether a rotation or an override gives the shift


`rotation` A rotation layer's turn.
`override` An override someone added.

Enum values:

rotation

override

rotation_id

string | null · uuid · required

The rotation layer, for rotation shifts

override_id

string | null · uuid · required

The override, for override shifts

OnCallShiftSource

string · enum

Enum values:

rotation

override

Where an on-call shift comes from.


`rotation` A rotation layer's turn.
`override` An override someone added.

OnCallShiftUnit

string · enum

Enum values:

hours

days

weeks

The unit of a rotation's shift length.


`hours` Absolute hours: shifts keep their length across DST changes.
`days` Calendar days in the schedule's time zone: the handoff stays at the same wall-clock time.
`weeks` Calendar weeks in the schedule's time zone.

OnCallTimeline

Who is on call over a window: each layer's shifts, the overrides, and the result.

from

string · date-time · required

Start of the window

to

string · date-time · required

End of the window

timezone

string · required

The schedule's time zone

Example: Europe/Amsterdam

OnCallTimelineLayer[] · required

Each layer's shifts, top layer first

OnCallShift[] · required

Override shifts

OnCallShift[] · required

Who actually gets paged when: overrides first, then the highest layer. Gaps are absent

OnCallTimelineLayer

One rotation layer's shifts in a timeline.

rotation_id

string · uuid · required

Rotation ID

name

string · required

Rotation name

Example: Weekly primary

position

integer · required

Layer order; 0 is the top layer

OnCallShift[] · required

The layer's own shifts, before overrides and higher layers

Plan

string · enum

Enum values:

none

starter

team

business

Billing plans (docs/billing.md). The plan sets the features; the number of monitors + hosts is the subscription's quantity, priced per monitor (config/billing.php). None is a team without a paid plan: on the trial (which uses the trial plan's limits) or unpaid (paused).


`none` No paid plan and no trial: monitoring is paused until the team subscribes.
`starter` Starter: up to 3 members and 1 status page, the team default escalation path only.
`team` Team: unlimited members, 5 status pages, on-call, more escalation paths and private status pages.
`business` Business: everything in Team, with unlimited status pages and more subscribers and notification credits.

Project

A project: a group of monitors and hosts that share escalation paths. Every team has one default project, which can't be deleted.

id

string · uuid · required

Project ID

name

string · required

Project name

Example: Production

is_default

boolean · required

Whether this is the team's default project, which new monitors and hosts join when no project is given. It can't be deleted

EscalationPathReference[] · required

Escalation paths that alerts of this project's monitors and hosts follow, unless a monitor or host sets its own

created_at

string · date-time · required

When the project was created

ProjectReference

A project a resource belongs to or applies to.

id

string · uuid · required

Project ID

name

string · required

Project name

Example: Production

PublishStatusPageUpdate

An update of the team's incident to publish on a status page.

status_page_id

string · uuid · required

The status page to publish on

status

IncidentStatus · enum · required

The public incident's status with this update. resolved closes every open impact period


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

body

string · required

The message visitors read, in Markdown

Example: We found the cause and are rolling out a fix.

impact

ImpactLevel · enum

The public incident's own impact, used while it affects no components; otherwise it follows the components


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

published_at

string | null · date-time

When it happened, now or in the past (default now)

notify_subscribers

boolean

Email confirmed subscribers about the update (default true). false changes the page without emails

ResolveIncident

How an incident ends.

message

string | null

A closing update, in Markdown; leave out for none

Example: Fixed by rolling back release 412.

ResourcePauseReason

string · enum

Enum values:

plan_limit

team_paused

Why the system (not the user) paused a monitor, host, domain or status page.


`plan_limit` Over the plan limit (trial expired, downgrade); the oldest resources stay active.
`team_paused` The whole team's monitoring is paused (no paid plan, staff action or scheduled deletion).

SavePostmortem

An incident's postmortem.

content

string · required

The whole postmortem, in Markdown; replaces the current one

published

boolean

Mark it as published (final), or as a draft again. Leave out to keep its state

SaveStatusPageIncidentWriteup

The public write-up to save on a status page incident.

published

boolean · required

true shows it on the public incident page; false keeps it as a draft, or takes a published write-up down

content

string | null · maxLength: 50000

The write-up, in Markdown, at most 50,000 characters. Required to publish

Example: On 26 October, card payments failed for 18 minutes after a configuration change.

notify_subscribers

boolean

Email confirmed subscribers when this publishes the write-up for the first time (default false). Edits to a published write-up are never emailed

ScheduledMaintenance

Planned maintenance announced on a status page. While it is in progress its components show as under maintenance.

id

string · uuid · required

Maintenance ID

title

string · required

Title shown on the page

Example: Database upgrade

description

string | null · required

What happens and what visitors notice, in Markdown; null for none

Example: The API is read-only for about ten minutes.

status

ScheduledMaintenanceStatus · enum · required

Where it stands


`scheduled` Announced, not started yet.
`in_progress` Going on: its components show as under maintenance.
`completed` Over.

Enum values:

scheduled

in_progress

completed

automatic

boolean · required

true: starts at scheduled_at and completes at scheduled_until by itself (without a planned end it is completed by hand). false: only changes when you set status

scheduled_at

string · date-time · required

Planned start

scheduled_until

string | null · date-time · required

Planned end; null for none

started_at

string | null · date-time · required

When it actually went in progress; null before

completed_at

string | null · date-time · required

When it completed; null before

StatusPageAffectedComponent[] · required

Components it affects

created_at

string · date-time · required

When it was created

ScheduledMaintenanceStatus

string · enum

Enum values:

scheduled

in_progress

completed

Where a scheduled maintenance stands.


`scheduled` Announced, not started yet.
`in_progress` Going on: its components show as under maintenance.
`completed` Over.

Severity

string · enum

Enum values:

critical

high

medium

low

How urgent an alert or incident is. Escalation paths can branch on it.


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

StatusPage

A status page: the public page that shows your components' status, incidents and maintenance.

id

string · uuid · required

Status page ID

name

string · required

Name, shown as the page title

Example: Acme status

slug

string · required

The page's subdomain under spectate.page

Example: acme

visibility

StatusPageVisibility · enum · required

Who can open the page


`public` Anyone.
`private` Visitors with the access link, or the password when the page has one. Part of the Team plan and up.

Enum values:

public

private

status

ComponentStatus · enum · required

The headline status visitors see: the worst status of the components, and the impact of open incidents that affect no components


`operational` Working normally.
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.
`under_maintenance` In planned maintenance.

Enum values:

operational

degraded_performance

partial_outage

major_outage

under_maintenance

public_url

string · uri · required

Where visitors open the page: the custom domain once it is verified, else the spectate.page address

Example: https://acme.spectate.page

access_url

string · uri · required

Opens the page while it is private. Anyone with it gets in; rotateStatusPageAccessKey replaces it

custom_domain

string | null · required

Your own domain for the page; null for none. Visitors use it once it is verified

Example: status.acme.com

required

password_protected

boolean · required

Whether visitors of a private page can also get in with the password

has_password

boolean · required

Whether the page has a password. The password itself is never returned

language

StatusPageLanguage · enum · required

Language of the public page and of subscriber emails


`en` English.
`nl` Dutch.
`de` German.
`fr` French.
`es` Spanish.
`it` Italian.

Enum values:

match_browser_language

boolean · required

Visitors get their browser's language when it is supported, and can switch languages in the footer

StatusPageTheme · required

How the page looks and what it shows

collapsed_groups

string[] · required

Component groups that start collapsed

automation_enabled

boolean · required

Monitor automation: when a monitor linked to a component goes down, an incident "{component} is down" (major outage) opens with automation_down_message, and resolves with automation_recovered_message when the monitor recovers

automation_down_message

string | null · required

Posted when a linked monitor goes down; {component} and {status_page} are filled in. null for a default message

Example: We're looking into problems with {component}.

automation_recovered_message

string | null · required

Posted when the monitor recovers; null for a default message

Example: {component} is working again.

issue_reports_enabled

boolean · required

Visitors can report problems from the page

issue_reports_show_on_page

boolean · required

While reports are above the threshold, the page says that users are reporting issues

issue_reports_alerts_enabled

boolean · required

Reaching the threshold fires an alert

issue_reports_threshold

integer · required

Reports within issue_reports_window_minutes that count as above the threshold, 1 to 1000

issue_reports_window_minutes

integer · required

The window for the threshold, in minutes, 1 to 1440

issue_reports_triggered_at

string | null · date-time · required

Since when reports are above the threshold; null when they aren't

traffic_alerts_enabled

boolean · required

Alert when views reach traffic_alert_multiplier times the usual views for the time of day

traffic_alert_multiplier

integer · required

How many times the usual views alert, 2 to 100

traffic_alert_min_views

integer · required

Fewer views than this in 15 minutes never alert, 10 to 1000000

EscalationPathReference[] · required

Escalation paths the page's alerts (issue reports and traffic spikes) follow; empty for the team's default path

required

created_at

string · date-time · required

When the page was created

StatusPageAffectedComponent

A component an incident or maintenance affects, and how badly.

StatusPageComponentReference · required

The component

impact

ImpactLevel · enum · required

For an incident: the ongoing impact, else the worst it had. For a maintenance: the impact it announces


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

StatusPageAffectedComponentInput

A component an incident or maintenance affects, and how badly.

component_id

string · uuid · required

A component of the status page

impact

ImpactLevel · enum · required

How badly; none for not affected


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

StatusPageAffectedComponentInputRequest

A component an incident or maintenance affects, and how badly.

component_id

string · uuid · required

A component of the status page

impact

ImpactLevel · enum · required

How badly; none for not affected


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

StatusPageComponent

A component of a status page: a part of your service visitors see the status of, optionally following a monitor.

id

string · uuid · required

Component ID

name

string · required

Name shown on the page

Example: API

description

string | null · required

Text shown with the name; null for none

Example: Our public REST API.

group

string | null · required

Group the component is shown in; null for the page's default group

Example: Backend

position

integer · required

Order within the group, lowest first

status

ComponentStatus · enum · required

The status, derived and read-only: the worst ongoing impact of an unresolved incident, then under_maintenance during maintenance, else operational


`operational` Working normally.
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.
`under_maintenance` In planned maintenance.

Enum values:

operational

degraded_performance

partial_outage

major_outage

under_maintenance

required

show_latency_chart

boolean · required

Whether the page shows the monitor's response time chart under the component

created_at

string · date-time · required

When the component was created

StatusPageComponentReference

A component of the status page.

id

string · uuid · required

Component ID

name

string · required

Component name

Example: API

StatusPageCustomDomainCheck

The result of looking up a custom domain's DNS records.

domain

string · required

The domain

Example: status.example.com

verified

boolean · required

Whether ownership is proven, now or earlier. A proven domain stays verified when a later check fails

verified_at

string | null · date-time · required

When ownership was proven; null before

StatusPageDnsRecordCheck[] · required

The records that can verify the domain; one passing record is enough

routed

boolean · required

Whether visitors reach the page: the domain points at it by CNAME or resolves to the same addresses. The page and its certificate need this; the certificate is issued on the first visit

addresses

string[] · required

Addresses the domain resolves to (A and AAAA records)

checked_at

string · date-time · required

When DNS was checked

StatusPageCustomDomainVerification

How to prove you own a status page's custom domain: create the TXT record, or point the domain at the CNAME target (which also routes visitors to the page). Then call `verifyStatusPageCustomDomain`.

verified

boolean · required

Whether ownership is proven. A proven domain stays verified when a later check fails

verified_at

string | null · date-time · required

When ownership was proven; null before

txt_record_name

string · required

Name of the TXT record to create

Example: _spectate.status.example.com

txt_record_value

string · required

Value of the TXT record

Example: spectate-verify=4f1c0c2e9b7a4d3f8e6a5b2c1d0e9f8a

cname_target

string · required

Host to point a CNAME at (subdomains), or an ALIAS (root domains); unique to the page

Example: 4f1c0c2e9b7a4d3f.custom.spectate.page

StatusPageDisplay

What a status page shows.

show_uptime_history

boolean · required

Show each component's uptime history

show_latency_charts

boolean · required

Show response time charts under components that have them on

show_incidents

boolean · required

Show incidents

show_maintenances

boolean · required

Show scheduled maintenance

show_subscriptions

boolean · required

Let visitors subscribe to updates by email (when the plan includes subscribers)

history_layout

StatusPageHistoryLayout · enum · required

How the uptime history is laid out


`calendar` A calendar of days (default).
`timeline` One row of bars per component.

Enum values:

calendar

timeline

uptime_bars

StatusPageUptimeBars · enum · required

How a day with problems is drawn in the uptime history


`fill` Colour part of the bar, by how long the problem lasted (default).
`solid` Colour the whole bar.

Enum values:

fill

solid

StatusPageDisplayInput

What a status page shows; settings left out keep their value.

show_uptime_history

boolean

Show each component's uptime history

show_latency_charts

boolean

Show response time charts under components that have them on

show_incidents

boolean

Show incidents

show_maintenances

boolean

Show scheduled maintenance

show_subscriptions

boolean

Let visitors subscribe to updates by email (when the plan includes subscribers)

history_layout

StatusPageHistoryLayout · enum

How the uptime history is laid out


`calendar` A calendar of days (default).
`timeline` One row of bars per component.

Enum values:

calendar

timeline

uptime_bars

StatusPageUptimeBars · enum

How a day with problems is drawn in the uptime history


`fill` Colour part of the bar, by how long the problem lasted (default).
`solid` Colour the whole bar.

Enum values:

fill

solid

StatusPageDnsRecordCheck

One DNS record that can verify a custom domain, with what DNS returned for it.

type

DnsRecordType · enum · required

Record type: CNAME or TXT

Enum values:

AAAA

CNAME

TXT

CAA

SOA

name

string · required

Record name

Example: _spectate.status.example.com

expected

string · required

The value it needs

Example: spectate-verify=4f1c0c2e9b7a4d3f8e6a5b2c1d0e9f8a

found

string[] · required

The values DNS returned; empty when there is no such record

ok

boolean · required

Whether this record passes

StatusPageFont

string · enum

Enum values:

geist

inter

ibm-plex-sans

dm-sans

manrope

space-grotesk

source-serif

system

Typefaces of a status page.


`geist` Geist (default).
`inter` Inter.
`ibm-plex-sans` IBM Plex Sans.
`dm-sans` DM Sans.
`manrope` Manrope.
`space-grotesk` Space Grotesk.
`source-serif` Source Serif.
`system` The visitor's system font.

StatusPageHistoryLayout

string · enum

Enum values:

calendar

timeline

How a status page shows the uptime history of its components.


`calendar` A calendar of days (default).
`timeline` One row of bars per component.

StatusPageImpactPeriod

A period during which an incident affected a component. Periods drive the component's status and its uptime history.

id

string · uuid · required

Period ID

StatusPageComponentReference · required

The affected component

impact

ComponentImpact · enum · required

How badly


`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

degraded_performance

partial_outage

major_outage

started_at

string · date-time · required

When it started

ended_at

string | null · date-time · required

When it ended; null while it goes on

duration_seconds

integer · required

How long it lasted so far, in seconds

StatusPageIncident

A public incident on a status page: what visitors read, separate from the team's own incident (which can publish to it). Its impact on components is a timeline of periods.

id

string · uuid · required

Status page incident ID

title

string · required

Title shown on the page

Example: Elevated API error rates

status

IncidentStatus · enum · required

Where it stands: the status of its latest update


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

impact

ImpactLevel · enum · required

The worst ongoing impact of its periods (the peak once resolved), none without one. An incident that affects no components keeps the impact it was given


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage

incident_id

string | null · uuid · required

The team's incident that publishes to it; null for none

automated

boolean · required

Opened by monitor automation and still managed by it, until someone posts an update or edits its timeline

StatusPageIncidentUpdate[] · required

Updates, newest first

StatusPageAffectedComponent[] · required

Components it affects or affected

StatusPageImpactPeriod[] · required

The impact timeline, oldest first

resolved_at

string | null · date-time · required

When it was resolved; null while open

created_at

string · date-time · required

When it was created

required

StatusPageIncidentUpdate

An update of a status page incident, as visitors read it.

id

string · uuid · required

Update ID

status

IncidentStatus · enum · required

The incident's status with this update


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

body

string · required

The message, in Markdown

Example: We found the cause and are rolling out a fix.

published_at

string · date-time · required

When it happened; updates are ordered by it

created_at

string · date-time · required

When it was written

StatusPageIncidentWriteup

A public write-up of a status page incident: what happened and what changes, for visitors. Published, it shows on the public incident page; otherwise it's a draft only the team sees.

content

string · required

The write-up, in Markdown

Example: On 26 October, card payments failed for 18 minutes after a configuration change.

published

boolean · required

Whether visitors see it on the public incident page

published_at

string | null · date-time · required

When it was first published; null while it's a draft. Editing a published write-up keeps this time

StatusPageIssueReport

A problem a visitor reported from the public page.

id

string · uuid · required

Report ID

message

string | null · required

What the visitor wrote; null for nothing

Example: Checkout keeps timing out.

email

string | null · email · required

The address the visitor left for a reply. Only returned to tokens with status-pages:write; null for others (see has_email) and when they left none

Example: jane@example.com

has_email

boolean · required

Whether the visitor left an email address, also when it isn't returned

StatusPageComponentReference[] · required

Components the visitor picked; deleted components are left out

created_at

string · date-time · required

When it was reported

StatusPageLanguage

string · enum

Enum values:

Languages of the public status page chrome and the subscriber emails. Customer-written content (names, incident titles and updates) is never translated.


`en` English.
`nl` Dutch.
`de` German.
`fr` French.
`es` Spanish.
`it` Italian.

StatusPageSubscriber

Someone who subscribed to a status page's updates by email. Only confirmed subscribers (double opt-in) get emails.

id

string · uuid · required

Subscriber ID

email

string · email · required

Email address

Example: jane@example.com

confirmed_at

string | null · date-time · required

When they confirmed the subscription; null while it awaits confirmation

language

StatusPageLanguage · enum · required

Language of their emails: the page's language when they subscribed, or the one they read the page in


`en` English.
`nl` Dutch.
`de` German.
`fr` French.
`es` Spanish.
`it` Italian.

Enum values:

created_at

string · date-time · required

When they subscribed

StatusPageTheme

How a status page looks. `null` values use the default.

accent

string | null · required

Brand colour, as hex; null for Spectate's yellow (#ffea00)

Example: #0f766e

required

StatusPageThemeColors · required

Colours overriding the preset in light mode

StatusPageThemeColors · required

Colours overriding the preset in dark mode

StatusPageDisplay · required

What the page shows

logo_url

string | null · uri · required

The logo, uploaded in the dashboard; null for none (the page shows its name)

logo_dark_url

string | null · uri · required

The logo for dark mode; null shows logo_url in dark mode too

favicon_url

string | null · uri · required

The favicon, uploaded in the dashboard; null for Spectate's

StatusPageThemeColors

Colours that override the preset in one colour scheme, as hex (`#0f766e`). `null` keeps the preset's colour.

background

string | null · required

Page background

Example: #ffffff

surface

string | null · required

Background of cards and panels

Example: #f8fafc

foreground

string | null · required

Text

Example: #0f172a

muted

string | null · required

Secondary text

Example: #64748b

border

string | null · required

Lines and borders

Example: #e2e8f0

success

string | null · required

Operational

Example: #16a34a

warning

string | null · required

Degraded performance and maintenance

Example: #ca8a04

partial

string | null · required

Partial outage

Example: #ea580c

danger

string | null · required

Major outage

Example: #dc2626

info

string | null · required

Information, such as scheduled maintenance

Example: #2563eb

StatusPageThemeColorsInput

Colours that override the preset in one colour scheme, as hex (`#0f766e` or `#0f7`). Colours left out keep their value; `null` goes back to the preset's colour.

background

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Page background

Example: #ffffff

surface

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Background of cards and panels

Example: #f8fafc

foreground

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Text

Example: #0f172a

muted

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Secondary text

Example: #64748b

border

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Lines and borders

Example: #e2e8f0

success

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Operational

Example: #16a34a

warning

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Degraded performance and maintenance

Example: #ca8a04

partial

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Partial outage

Example: #ea580c

danger

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Major outage

Example: #dc2626

info

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Information, such as scheduled maintenance

Example: #2563eb

StatusPageThemeInput

Changes to how a status page looks. Settings left out keep their value, also inside `light`, `dark` and `display`; `null` goes back to the default. Logos and the favicon are uploaded in the dashboard.

accent

string | null · pattern: ^#(?:[0-9a-f]{3}|[0-…

Brand colour, as hex; null for Spectate's yellow

Example: #0f766e

StatusPageDisplayInput

What the page shows

StatusPageThemeMode

string · enum

Enum values:

light

dark

system

The colour scheme visitors of a status page get before they pick one.


`light` Light.
`dark` Dark.
`system` The visitor's operating system setting.

StatusPageThemePreset

string · enum

Enum values:

spectate

paper

graphite

ocean

forest

lavender

ember

contrast

Colour presets of a status page. Colours set in light and dark override the preset.


`spectate` Neutral greys (default).
`paper` Warm cream and sepia ink.
`graphite` Soft charcoal, strong lines.
`ocean` Clear blues, navy in dark mode.
`forest` Sage greens, deep pine in dark mode.
`lavender` Lilac, aubergine in dark mode.
`ember` Peach and terracotta, warm in dark mode.
`contrast` High contrast: pure black and white, bold lines.

StatusPageUptimeBars

string · enum

Enum values:

fill

solid

How a day with problems is drawn in a status page's uptime history.


`fill` Colour part of the bar, by how long the problem lasted (default).
`solid` Colour the whole bar.

StatusPageVisibility

string · enum

Enum values:

public

private

Who can open a status page.


`public` Anyone.
`private` Visitors with the access link, or the password when the page has one. Part of the Team plan and up.

Subscription

The team's paid subscription.

plan

Plan · enum · required

The plan paid for


`none` No paid plan and no trial: monitoring is paused until the team subscribes.
`starter` Starter: up to 3 members and 1 status page, the team default escalation path only.
`team` Team: unlimited members, 5 status pages, on-call, more escalation paths and private status pages.
`business` Business: everything in Team, with unlimited status pages and more subscribers and notification credits.

Enum values:

none

starter

team

business

interval

BillingInterval · enum · required

How often it's billed


`month` Billed every month.
`year` Billed every year, for the price of 10 months.

Enum values:

month

year

quantity

integer · required

Monitors and hosts paid for: the team's limit of both together

status

SubscriptionStatus · enum · required

Where it stands at the payment provider


`created` Created; the first payment hasn't gone through yet.
`trial` In a trial period.
`active` Paid and running.
`on_grace_period` Cancelled, running until the end of the paid period.
`paused` Paused.
`canceled` Ended.

Enum values:

created

trial

active

on_grace_period

paused

canceled

cancelled

boolean · required

Whether it's cancelled and runs until current_period_ends_at; it can be resumed in the dashboard until then

renews_at

string | null · date-time · required

When it renews; null when it won't

current_period_ends_at

string | null · date-time · required

End of the period paid for

payment_method

string | null · required

How it's paid; null when unknown

Example: Card ending in 4242

SubscriptionStatus

string · enum

Enum values:

created

trial

active

on_grace_period

paused

canceled

Where a subscription stands at the payment provider.


`created` Created; the first payment hasn't gone through yet.
`trial` In a trial period.
`active` Paid and running.
`on_grace_period` Cancelled, running until the end of the paid period.
`paused` Paused.
`canceled` Ended.

Team

The team the API token belongs to.

id

string · uuid · required

Team ID

name

string · required

Name

Example: Acme

email

string | null · email · required

The team's billing email: Spectate gives it to Vatly, which sends the invoices and receipts, at the team's first checkout (a plan or credits). null to use the address of the member who starts that checkout

Example: ops@acme.com

timezone

string · required

The team's time zone; escalation path working hours without a time zone of their own use it

Example: Europe/Amsterdam

plan

Plan · enum · required

The plan whose limits apply now: the trial plan during the trial


`none` No paid plan and no trial: monitoring is paused until the team subscribes.
`starter` Starter: up to 3 members and 1 status page, the team default escalation path only.
`team` Team: unlimited members, 5 status pages, on-call, more escalation paths and private status pages.
`business` Business: everything in Team, with unlimited status pages and more subscribers and notification credits.

Enum values:

none

starter

team

business

on_trial

boolean · required

Whether the team is on its trial, without a paid plan yet

trial_ends_at

string | null · date-time · required

When the trial ends; null without one

monitoring_paused_at

string | null · date-time · required

Since when all monitoring is paused; null while it runs

required

deletion_scheduled_at

string | null · date-time · required

When the team and all its data will be deleted; null when it isn't scheduled. An owner can cancel it in the dashboard

required

created_at

string · date-time · required

When the team was created

TeamDeletionReason

string · enum

Enum values:

requested

unpaid


`requested` An owner deleted the team.
`unpaid` Monitoring was paused for too long without a paid plan (`team_lifecycle.unpaid`).

TeamPauseReason

string · enum

Enum values:

unpaid

deletion_scheduled

staff


`unpaid` No paid plan: the trial or the subscription ended. Resumes when the team subscribes.
`deletion_scheduled` The team is scheduled for deletion.
`staff` Paused by Spectate staff from Overwatch (abuse, payment issues, on request).

TeamRole

string · enum

Enum values:

owner

manager

admin

member

billing

A member's role in the team: what they may do.


`owner` Everything, including billing and deleting the team. One per team.
`manager` Everything but billing and deleting the team.
`admin` Configure monitoring, alerting, incidents and status pages; can't manage members or API tokens.
`member` Read everything; manage monitors and incidents, acknowledge alerts.
`billing` Billing and support only.

TestChannel

Options of a channel test.

enable

boolean

Enable the channel again when the test succeeds, also after Spectate turned it off. Default false

TlsMode

string · enum

Enum values:

none

tls

starttls

How TCP and mail monitors use TLS: not at all, from the first byte (implicit TLS, e.g. 465/993/995), or by upgrading a plaintext session with STARTTLS (SMTP), STARTTLS (IMAP) or STLS (POP3).


`none` Plain TCP, no TLS.
`tls` TLS from the first byte (implicit TLS), as on ports 465, 993 and 995.
`starttls` Upgrade a plaintext session: STARTTLS (SMTP, IMAP) or STLS (POP3). Mail monitors only.

UpdateAlertSource

Changes to an alert source; fields you leave out keep their value. Its type can't change.

name

string · maxLength: 255

Name shown on its alerts

Example: Production Alertmanager

enabled

boolean

Whether requests fire alerts

default_severity

Severity · enum

Severity of alerts sent without one


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

auto_resolve_minutes

integer | null · min: 5 · max: 10080

Minutes, 5 to 10080, after which an alert the tool stopped repeating resolves; null keeps it until resolved

secret

string | null · maxLength: 255

Signing secret; an empty string or null removes it. Write-only

AlertSourceSettingsInput

Type-specific settings; settings you leave out keep their value

project_id

string | null · uuid

Project whose escalation paths apply; null for none

escalation_path_ids

string[] · unique

Escalation paths for the source's alerts; replaces the current ones

UpdateChannel

Changes to a channel; fields you leave out keep their value. A channel's type can't change. Its settings object, when you send it, replaces the current settings, secrets included.

name

string · maxLength: 255

Name shown in escalation paths

Example: On-call email

enabled

boolean

Turn delivery on or off. Turning it on clears an automatic disable and its failure count

ChannelEmail

New settings of an email channel

ChannelSlackInput

New settings of a slack channel

New settings of a discord channel

New settings of a microsoft_teams channel

ChannelWebhookInput

New settings of a webhook channel

ChannelPushoverInput

New settings of a pushover channel

New settings of an sms channel

New settings of a voice channel

UpdateDomain

Changes to a monitored domain; fields you leave out keep their value.

name

string · maxLength: 255

Another domain to monitor instead; its registry data is looked up again

Example: example.org

enabled

boolean

Pause (false) or resume (true) monitoring. Resuming goes through the plan's domain limit

expiry_threshold_days

integer · min: 1 · max: 365

Days before expiry, 1 to 365, that the "expires soon" alert fires

alert_on_registry_changes

boolean

Raise an alert when the registrar, nameservers or DNSSEC keys change

project_id

string | null · uuid

Project of the domain; null for none

escalation_path_ids

string[]

Escalation paths for the domain's alerts; replaces the current ones

UpdateEscalationPath

Changes to an escalation path; fields you leave out keep their value. `steps` replaces all steps. Alerts already escalating keep the steps they started with.

name

string · maxLength: 255

Name, up to 255 characters

Example: On-call engineers

description

string | null · maxLength: 1000

What the path is for, up to 1000 characters

enabled

boolean

Whether alerts follow the path

is_default

boolean

Make it the team's default path (the current default stops being one), or not

MonitorAdvancedSettingsInput

All steps, in order, up to 99 in total; replaces the current steps

UpdateHost

Changes to a host; fields you leave out keep their value.

label

string · maxLength: 255

Name shown in the dashboard and alerts

Example: web-1

no_data_alert_minutes

integer | null · min: 2 · max: 1440

Minutes without data, 2 to 1440, before the "no data received" alert fires; null turns it off. A firing alert ends at once when the host is within the new window

Example: 10

project_id

string | null · uuid

Project of the host; null for none

escalation_path_ids

string[]

Escalation paths for the host's alerts; replaces the current ones. [] follows the project's paths, else the team's default path

UpdateIncident

Changes to an incident; fields you leave out keep their value. To post what's known, use `createIncidentUpdate`; to close it, `resolveIncident`.

title

string · maxLength: 255

What's going on

Example: Checkout fails for some customers

summary

string | null

More about it, in Markdown

severity

Severity · enum

How bad it is. Changing it is announced in the incident's Slack channel


`critical` Customers are affected right now.
`high` Serious, needs attention soon.
`medium` Should be looked at.
`low` Informational.

Enum values:

critical

high

medium

low

status

IncidentStatus · enum

Where it stands, without posting an update


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

UpdateMember

A member's new role.

role

TeamRole · enum · required

The role; owner can't be given (422)


`owner` Everything, including billing and deleting the team. One per team.
`manager` Everything but billing and deleting the team.
`admin` Configure monitoring, alerting, incidents and status pages; can't manage members or API tokens.
`member` Read everything; manage monitors and incidents, acknowledge alerts.
`billing` Billing and support only.

Enum values:

owner

manager

admin

member

billing

UpdateMonitor

Changes to a monitor; fields you leave out keep their value. To pause or resume it, use `pauseMonitor` and `resumeMonitor`.

label

string · maxLength: 255

Name shown in the dashboard, alerts and status pages

Example: Website

protocol

MonitorProtocol · enum

What to check. Changing it requires uri in the new protocol's format


`https` An HTTPS request to the URL in `uri`, with a certificate check.
`http` An HTTP request to the URL in `uri`.
`icmp` A ping (ICMP echo) to the host in `uri`.
`tcp` A TCP connect to `host:port`, optionally with a TLS handshake and a banner keyword.
`dns` A DNS query for a record of the name in `uri`, optionally compared with expected values.
`smtp` An SMTP greeting from `host:port` (port optional), with STARTTLS by default.
`imap` An IMAP greeting from `host:port` (port optional), over TLS by default.
`pop3` A POP3 greeting from `host:port` (port optional), over TLS by default.

Enum values:

https

http

icmp

tcp

dns

smtp

imap

pop3

uri

string · maxLength: 2048

The target, in the format of the (new) protocol; see createMonitor. For http and https the scheme is optional and must match the protocol; it is stored without the scheme

Example: https://example.com/health

interval

MonitorInterval · enum

Seconds between checks. Plans set the shortest allowed interval

Enum values:

120

180

300

600

900

1800

primary_location

MonitorLocation · enum

Region the checks run from


`eu` Europe.
`us` North America.
`as` Asia.
`oc` Oceania.

Enum values:

timeout

integer · min: 1 · max: 60

Seconds before a check times out, 1 to 60

max_redirects

integer | null · min: 0 · max: 10

HTTP: redirects to follow, 0 to 10; null follows up to 10

MonitorHeader[]

HTTP: up to 25 headers to send; replaces the current headers

Protocol-specific settings; settings you leave out keep their value

allow_insecure_requests

boolean

HTTPS: accept invalid certificates for the check itself

expected_http_status_code

string | null · minLength: 1 · maxLength: 200

HTTP: status codes that count as up; null accepts 2xx and 3xx

certificate_check_enabled

boolean

Check the TLS certificate for validity and expiry

certificate_expiry_threshold_days

integer · min: 1 · max: 90

Days before the certificate expires that the "expires soon" alert fires, 1 to 90

performance_check_enabled

boolean

Raise alerts for slow responses

performance_hard_threshold

integer | null · min: 1 · max: 30000

Milliseconds, 1 to 30000: a response slower than this raises an alert

performance_deviation_percentage

integer | null · min: 1 · max: 500

Percent, 1 to 500: alert when recent response times are this much above the earlier ones

performance_deviation_timeframe

integer | null · min: 15 · max: 120

Minutes of recent checks to compare: 15 to 120 in steps of 15

performance_deviation_before_timeframe

integer | null · min: 60 · max: 1440

Minutes before those to compare with: 60 to 1440 in whole hours

regional_performance_enabled

boolean

project_id

string | null · uuid

Project of the monitor; null for none

escalation_path_ids

string[]

Escalation paths for the monitor's alerts; replaces the current ones. [] follows the project's paths, else the team's default path

UpdateOnCallRotation

Changes to a rotation layer; fields you leave out keep their value.

name

string · maxLength: 255

Name

Example: Weekly primary

position

integer · min: 0 · max: 100

Move the layer to this position, 0 for the top; the others shift

shift_unit

OnCallShiftUnit · enum

The unit of the shift length


`hours` Absolute hours: shifts keep their length across DST changes.
`days` Calendar days in the schedule's time zone: the handoff stays at the same wall-clock time.
`weeks` Calendar weeks in the schedule's time zone.

Enum values:

hours

days

weeks

shift_length

integer · min: 1 · max: 720

Shift length in shift_units

start_date

string · date · pattern: ^\d{4}-\d{2}-\d{2}$

The first handoff's date

handoff_time

string

Handoff time as HH:MM

Example: 09:00

participant_ids

string[] · minItems: 1 · maxItems: 50 · unique

Members taking turns, in order; replaces the current ones

UpdateOnCallSchedule

Changes to an on-call schedule; fields you leave out keep their value.

name

string · maxLength: 255

Name

Example: Primary

description

string | null · maxLength: 1000

What the schedule covers

timezone

string

IANA time zone. Rotations keep their dates and wall-clock handoff times in the new zone

Example: Europe/Amsterdam

UpdateProject

Changes to a project; fields left out keep their value.

name

string · maxLength: 255

Project name

Example: Production

escalation_path_ids

string[] · unique

Escalation paths for alerts of the project's monitors and hosts; replaces the current list. Send [] for none

UpdateScheduledMaintenance

Changes to a maintenance; fields left out keep their value.

title

string · maxLength: 255

Title shown on the page, up to 255 characters

Example: Database upgrade

description

string | null

What happens and what visitors notice, in Markdown; null for none

Example: The API is read-only for about ten minutes.

status

ScheduledMaintenanceStatus · enum

Move it along by hand: in_progress starts it, completed completes it (stamping completed_at), scheduled puts it back


`scheduled` Announced, not started yet.
`in_progress` Going on: its components show as under maintenance.
`completed` Over.

Enum values:

scheduled

in_progress

completed

automatic

boolean

true: starts and completes on schedule. false: only changes when you set status

scheduled_at

string · date-time

Planned start

scheduled_until

string | null · date-time

Planned end; null for none

completed_at

string | null · date-time

When it completed. Setting it completes the maintenance

Components it affects; replaces the current list. Components of the status page only

notify_subscribers

boolean

Email confirmed subscribers when this starts or completes the maintenance (default true)

UpdateStatusPage

Changes to a status page; settings left out keep their value.

name

string · maxLength: 255

Name, shown as the page title

Example: Acme status

slug

string · maxLength: 63 · pattern: ^[a-z0-9](?:[a-z0-9-…

The page's subdomain under spectate.page: lowercase letters, digits and dashes. The old address stops working

Example: acme

visibility

StatusPageVisibility · enum

Who can open the page. Private pages are part of the Team plan and up (403 PLAN_FEATURE_UNAVAILABLE)


`public` Anyone.
`private` Visitors with the access link, or the password when the page has one. Part of the Team plan and up.

Enum values:

public

private

custom_domain

string | null · maxLength: 253

Example: status.acme.com

password_protected

boolean

Let visitors of a private page also get in with a password. Part of paid plans (403 PLAN_FEATURE_UNAVAILABLE otherwise); turning it on needs a password unless the page has one

password

string | null · minLength: 8 · maxLength: 255

The password, 8 to 255 characters. Write-only; leave out or send null to keep the current one

Example: correct-horse-battery

language

StatusPageLanguage · enum

Language of the public page and of subscriber emails


`en` English.
`nl` Dutch.
`de` German.
`fr` French.
`es` Spanish.
`it` Italian.

Enum values:

match_browser_language

boolean

Give visitors their browser's language when it is supported

StatusPageThemeInput

How the page looks and what it shows

collapsed_groups

string[] · maxItems: 100 · unique

Component groups that start collapsed; replaces the current list

automation_enabled

boolean

Monitor automation: when a monitor linked to a component goes down, open an incident "{component} is down" (major outage), and resolve it when the monitor recovers

automation_down_message

string | null · maxLength: 5000

Posted when a linked monitor goes down, up to 5000 characters; {component} and {status_page} are filled in. null for a default message

Example: We're looking into problems with {component}.

automation_recovered_message

string | null · maxLength: 5000

Posted when the monitor recovers, up to 5000 characters; null for a default message

Example: {component} is working again.

issue_reports_enabled

boolean

Let visitors report problems from the page

issue_reports_show_on_page

boolean

While reports are above the threshold, say on the page that users are reporting issues

issue_reports_alerts_enabled

boolean

Fire an alert when reports reach the threshold

issue_reports_threshold

integer · min: 1 · max: 1000

Reports within issue_reports_window_minutes that count as above the threshold, 1 to 1000

issue_reports_window_minutes

integer · min: 1 · max: 1440

The window for the threshold, in minutes, 1 to 1440

traffic_alerts_enabled

boolean

Alert when views reach traffic_alert_multiplier times the usual views for the time of day

traffic_alert_multiplier

integer · min: 2 · max: 100

How many times the usual views alert, 2 to 100

traffic_alert_min_views

integer · min: 10 · max: 1000000

Fewer views than this in 15 minutes never alert, 10 to 1000000

escalation_path_ids

string[]

Escalation paths for the page's alerts (issue reports and traffic spikes); replaces the current ones. Empty for the team's default path

UpdateStatusPageComponent

Changes to a component; fields left out keep their value. Its status follows incidents and maintenance and can't be set.

name

string · maxLength: 255

Name shown on the page

Example: API

description

string | null

Text shown with the name; null for none

Example: Our public REST API.

group

string | null · maxLength: 255

Group to show the component in; null for the page's default group

Example: Backend

position

integer · min: 0

Order within the group, lowest first, from 0. New components go last when left out

monitor_id

string | null · uuid

A monitor of the team whose uptime history and response times the component shows, and whose downtime opens an incident when monitor automation is on; null for none

show_latency_chart

boolean

Show the monitor's response time chart under the component

UpdateStatusPageImpactPeriod

Changes to an impact period; fields left out keep their value.

component_id

string · uuid

A component of the status page

impact

ComponentImpact · enum

How badly


`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

degraded_performance

partial_outage

major_outage

started_at

string · date-time

When it started, now or in the past

ended_at

string | null · date-time

When it ended, after started_at and not in the future; null reopens the period

UpdateStatusPageIncident

Changes to a public incident without posting an update; fields left out keep their value.

title

string · maxLength: 255

Title shown on the page

Example: Elevated API error rates

status

IncidentStatus · enum

Where it stands. Changing it posts no update; resolved closes every open impact period now


`investigating` Declared; the cause isn't known yet.
`identified` Acknowledged: the cause is known and someone works on it.
`monitoring` A fix is out; watching whether it holds.
`resolved` Over.

Enum values:

investigating

identified

monitoring

resolved

impact

ImpactLevel · enum

The incident's own impact, used while it affects no components


`none` No impact (for a component: not affected).
`degraded_performance` Slower or less reliable than usual.
`partial_outage` Down for some visitors or some functions.
`major_outage` Down.

Enum values:

none

degraded_performance

partial_outage

major_outage