For: Anyone running a service customers depend on being up

Uptime & Incident Comms

A failed probe becomes a status-page update and a customer notice before support gets flooded with tickets.

$350/mo
+$590 one-time setup fee

What it watches

Uptime probes on your key endpoints

What it does

Updates your status page and notifies affected customers automatically when a probe fails repeatedly

When it asks a human first

Severity 2 or above pages your on-call channel immediately — the agent handles comms, never the actual fix to the underlying outage. A service that flaps — recovering and failing repeatedly within a short window — escalates as a single ongoing incident rather than paging fresh on every bounce, so on-call is not drowned in duplicate alerts for the same underlying problem.

How it works

  1. 1
    Probes your endpoints

    Runs continuous health checks against the URLs and services you specify. Checks run from multiple geographic regions by default, so a routing problem local to one network is never mistaken for your service actually being down everywhere.

  2. 2
    Confirms before alerting

    Waits for N consecutive failures (your threshold) before treating a blip as an incident — no alert fatigue from a single dropped check. The threshold is configurable per endpoint, not fixed globally, so a payment API you would flag on the second failure can sit next to a marketing page you are happy to give five.

  3. 3
    Updates the status page

    Posts an incident entry to your public status page automatically, so customers see it before they need to ask. The entry names only the affected component — checkout, API, dashboard — never a blanket "systems down" that overstates a partial outage into a full one.

  4. 4
    Notifies affected customers

    Sends a notice to accounts using the affected service, not a blast to your entire customer base. It cross-checks which accounts actually touched the affected component in the recent window, so a customer who never uses that feature does not get an alarming notice about downtime irrelevant to them.

  5. 5
    Closes the loop

    When the probe recovers, it posts a resolution update and a downtime summary — the fix itself is still on you. A recovery has to hold for a few consecutive clean checks before it posts the all-clear, so a service that flaps back up for thirty seconds and drops again is not falsely marked resolved.

What triggers it

  • · A probe on a monitored endpoint fails once
  • · A probe fails N consecutive times past your threshold
  • · A failed endpoint recovers
  • · An incident crosses your defined severity level

What you get

  • · A status-page update posted automatically at incident start and resolution
  • · A customer notice sent to affected accounts
  • · An internal alert to your on-call channel (Slack/Telegram)
  • · A post-incident summary with downtime duration

Not the right fit if

Not for static marketing sites with no uptime-sensitive service behind them — there is nothing here worth a status page. It’s also not the right layer if you’re already running a dedicated incident-management stack (PagerDuty, Statuspage, an SRE on-call rotation) — this agent is built to be the whole lightweight loop for a team that doesn’t have one yet, not a bolt-on next to an escalation system that already exists.

The math

At ~4 incidents/mo × 20 min, that is roughly 1.3 h/mo — an estimate, not a guarantee.

Frequently asked

No — it handles detection and communication. Someone on your team still fixes the actual problem.

As many as you list at setup — pricing scales with the number of monitored endpoints and check frequency.

Yes — internal alerts and customer comms can go to different channels simultaneously.

You define the thresholds at setup — typically a core service down vs. a degraded but functional one.

Yes — a response-time threshold runs alongside the pass/fail check, so a page that loads but takes ten seconds instead of one still triggers a degraded-performance incident, not just a hard outage.

Ready to put uptime & incident comms to work?

Tell us about your setup and we'll confirm it fits within 48 hours.

See the full workflow catalog