No alerts sent.
Nothing has been confirmed, so nobody is woken up.
Glowo checks your sites and APIs from up to 9 regions, pages your team only once a failure is confirmed, and keeps your status page current while you fix it. Or hand the day-to-day to Claude, Codex or Cursor.
No credit card. The Free plan never expires.
A simulated incident, check by check, from 9 regions. See exactly when Glowo pages your team, and every moment it deliberately does not.
Every region checks in fine. checkout.acme.dev is up.
One failed check. Glowo double-checks before it bothers anyone, so nobody is paged.
6 of 9 regions see errors at once. Glowo confirms with one more check first.
Confirmed from 6 of 9 regions. Your team gets one alert, and your status page opens an incident on its own.
Checks pass again. Glowo waits for a second pass, so a flicker cannot close the incident.
Recovery confirmed. Your team hears it was down for 2 min, 30 sec, and the incident closes itself.
Up · all 9 regions agree
No alerts sent.
Nothing has been confirmed, so nobody is woken up.
No incidents reported.
Checkout API is downInvestigating
Checkout API is downResolved
Glowo works with Claude Code, Codex, Cursor and any MCP client. Ask in plain words, and your agent sets up monitors, wires alerts, finds what is down and keeps your users informed, with only the access you approve.
Watch checkout.acme.dev from Europe and the US, and tell #ops if it goes down.
glowo_probe_urlglowo_create_notification_channelglowo_test_notification_channelglowo_create_monitorDone. Checkout API is checked from Gravelines, Paris, Ashburn, and #ops hears about any confirmed outage.
Is anything down? If so, acknowledge it and tell our users we are on it.
glowo_list_monitorsglowo_get_monitorglowo_acknowledge_monitorglowo_list_incidentsglowo_add_incident_updateCheckout API has been down since 03:18. I acknowledged the alert and told your users on status.acme.dev that you are on it.
We deploy checkout tonight at 22:00 for half an hour. Do not page anyone for it.
glowo_create_maintenance_windowScheduled. Checks keep running, nobody is paged during the window, and your status page shows the maintenance.
No agent can ever touch billing, members or workspace settings.
claude mcp add --scope user --transport http glowo https://api.glowo.dev/mcpcodex mcp add glowo --url https://api.glowo.dev/mcpOr point any MCP client at https://api.glowo.dev/mcp. Agent guide
Checks, alerts and a status page that work as one, so the first thing anyone hears about an outage is the truth.
Slack, email, SMS, PagerDuty, Discord, Telegram or your own webhook. Glowo confirms an outage before it pages anyone, so your team can trust every alert it gets.
How alerting worksYour brand, and your own domain on Pro and Business. Incidents open and close with your monitors, so customers know what is happening before they have to ask.
How status pages workInvite your team, share monitors and incidents, and decide who can change what. SAML SSO on Business.
Roles and permissionsOur own services, checked by the product you would use, straight from status.glowo.dev: the same page our customers read.
From status.glowo.dev
No per-check fees and no overage bills. Pay yearly and get two months free. Prices are in US dollars and exclude VAT.
For side projects and getting started.
$0/ month
Free forever, no card required
Start freeFor growing teams that need fast checks and SMS alerts.
$29/ month
or $290 a year
Start with ProFor teams with compliance, scale, and SSO needs.
$99/ month
or $990 a year
Start with BusinessPick a plan and pay at sign-up, or start on Free and upgrade later. Cancel in one click from the billing page.
Something else? Ask us or read the documentation.
HTTP(S) with assertions on status, body, headers and latency; TCP ports; ICMP ping; DNS records, including an alert on any unplanned change to an answer; WebSocket; gRPC health; and push monitors for cron jobs and private services, which call Glowo instead of being polled.
A single failed request is evidence, not an outage. Glowo waits for a failure to repeat over consecutive checks before it alerts, and a monitor that checks from several regions must fail from more than one of them. A service flapping between up and down pages you once, not forty times. Recovery works the same way: a monitor is back up when the checks say so, not when a timer expires.
In the EU. Glowo is operated from France, and accounts, monitors, check history and status pages live on servers in Germany and France. The checkers that run your monitors are distributed globally so checks happen close to your users, but they hold no customer data: they receive a job and return a result.
There is nothing to expire. The Free plan is free for as long as you want it, with 10 monitors, real alerts and a real status page, and no card to enter. Upgrade if and when you outgrow it.
Yes, from the billing page, in one click. You keep access until the end of the period you paid for and are never charged again. No retention call, no email to a sales team.
Prices are in US dollars and exclude VAT. Stripe adds the VAT of your billing country on top. An EU business can enter its VAT number at the first checkout to be reverse-charged, and add or change it later in the billing portal. Invoices are emailed automatically and downloadable there too.
Nothing is cut off mid-incident and nothing is billed as an overage. You get a warning at 80% of a limit and, at 100%, creating new resources is blocked until you upgrade or remove something. Existing monitors keep running.
Everything the dashboard does, the API does. There is an OpenAPI 3.1 specification, scoped API keys you create yourself, and a Model Context Protocol server with OAuth 2.1, so an agent such as Claude Code or Cursor can read status, create monitors and manage incidents with only the permissions you grant.
Add a URL and the checks start. The Free plan watches up to 10 monitors, with real alerts and a real status page, and never expires.