Keep a pulse on
everything
you ship.
Crons, queues, backups, deploys, APIs, whole sites — if it can ping us or be pinged, Holter watches it and tells you the moment a check catches it. Peace of mind that everything you built is still running.
Free for 5 monitors. No credit card. Watch your whole stack from one place.
Outbound checks · we ping them
Inbound heartbeats · they ping us
Uptime · last 34 days
Live demo — Holter detects, alerts, and recovers on its own.
Finding out you're down from an
angry customer is the worst way to find out.
The status page is green. The dashboards look fine. Then a support ticket comes in — and another — and your Slack lights up asking what's happening. Holter closes that gap. You hear it from an alert, not a support thread.
Users discover the outage. You discover it from them.
Three tools, four dashboards, and you still find out last.
Detected from outside, your team alerted on Slack, status page auto-updated.
Watch what you depend on
Business plan
When a vendor you rely on goes down,
hear it from us — not your customers.
Your app is only as up as the services behind it — Stripe, your database host, the auth provider, that one API everything quietly depends on. Holter watches their public status pages and alerts your team the moment one reports degradation or an outage. So when checkout breaks, you already know it's them — before the support tickets start.
24 popular vendors ready to add — or paste any status-page URL. Re-checked every few minutes; alerts flow to the same channels as your monitors, and you can list them on your status page as “Upstream services.”
One product, three jobs
Monitoring, alerting, and status pages. Together.
No glue code between three vendors. Holter watches your endpoints, alerts the right channel, and tells your users — all from one timeline.
Response time · checkout-api
last 60 minIncident timeline
#INC-4821 · resolved-
14:02:11 Detected
checkout-api failing its health check.
-
14:02:23 Alerted
Slack, Discord & email notified.
-
14:05:08 Status page
Auto-posted "Investigating payment errors" to subscribers.
-
14:14:52 Resolved
Rolled back deploy. Recovery confirmed. MTTR 12m41s.
90-day uptime
99% operationalUp and watching in 3 minutes
Scan a site or add a monitor
Paste a URL and Holter probes it live and builds a page with the monitors it finds — or add one by hand and grab a heartbeat URL for a cron. Checks run on a schedule you choose, as often as every 60 seconds.
Choose where alerts go
Slack, Discord, email, or a webhook — route by tag, immediate or batched. (SMS & phone are on the roadmap.)
Publish a status page
Branded, on your domain, auto-updated during incidents. On Pro, optionally let visitors subscribe — you don't lift a finger.
Status pages
A status page that
updates itself.
Publish a branded status page tied to your real monitors — components, uptime bars, and response-time graphs update the moment an incident opens or resolves. No manual editing mid-outage.
- ✓ Per-component uptime and response-time graphs, live from your monitors
- ✓ Auto-posts incidents as they open and resolve
- ✓ Your own domain and optional visitor subscribers on Pro
- ✓ Embeddable status & uptime badges for your README or site
Live right now from Holter status
These are Holter's real current status badges. Create your own and embed it in your README, docs, footer, or customer portal.
status.holter.shAll systems operational
All 5 services are up and running.
90-day uptime
99.98%
Avg response
142ms
Operational
5/ 5
Response time
Now 142ms avg
New · MCP server
Or let your coding agent
set it all up.
Holter runs an MCP server. Point Claude Code, Cursor, or any MCP-capable agent at your repo — your agent reads your crons, queues, health endpoints and domains locally, proposes the monitors, and hands you the heartbeat code to drop in. Your agent applies only the monitors you approve.
- ✓ Your agent reads your codebase locally — nothing to wire up by hand
- ✓ Proposes the monitors and the code for you to review in your agent
- ✓ Works with a token, never your keys — it can't touch your systems
$ agent scanning repo…
✓ found
→ 4 monitors proposed · review & apply
Or just ask your agent
"Set up Holter monitoring for this repo — read holter.sh/use and follow it."
Holter vs. the alternatives
A cron script is free until it pages no one at 3am. An observability suite is powerful until your CFO sees the invoice.
| Capability | Holter | A free ping script | Enterprise observability suite |
|---|---|---|---|
| Detect downtime from outside | ✓ as often as every 60s | ~ one box, your cron | ✓ if configured |
| Alerts to Slack / Discord / email | ✓ included | ✗ none | + add-on / seat |
| Hosted status page | ✓ included | ✗ build it | + extra product |
| Set up by your AI agent (MCP) | ✓ included | ✗ none | ✗ none |
| Alert when a vendor you rely on is down | ✓ Business | ✗ none | ✗ none |
| Setup time | 3 minutes | an afternoon | weeks + a consultant |
| Team members | Free ✗ · Pro 3 · Business 15 | ✗ solo only | + per-seat fees |
| Starting price | $0 → $15 → $99/mo | $0 + your time | $$$$ / year |
Built on one principle
Monitor everything. Hand over nothing.
No server agent, no credentials, no SDK in your code. Holter watches from the outside and lets the rest ping in.
Observe from outside
We fetch your URLs, read public DNS, and check certificates the way your users reach you — no access to your servers.
You push the proof
Crons, queues and backups send a heartbeat when they finish. You hand us a ping, never the keys.
The alert is the product
Quiet when things are fine, loud when they are not. Detect fast, page the right person, get out of the way.
Simple, transparent pricing
Start free with 5 monitors — no card. Every paid plan comes with a 14-day free trial.
Free
For side projects & first checks.
- ✓ Up to 5 monitors
- ✓ Dead-man heartbeats (minutes to monthly)
- ✓ Active checks every 5 minutes
- ✓ Email alerts
- ✓ 1 status page · 30-day uptime display
Pro
For teams running real production.
- ✓ Everything in Free
- ✓ Up to 100 monitors
- ✓ Active checks every 1 minute
- ✓ Email, Slack, Discord & webhook alerts
- ✓ Global checks — up to 5 probe locations
- ✓ 10 status pages · 90-day uptime display
- ✓ Custom-domain status pages
- ✓ 3 team members
Business
For global checks, dependencies & growing teams.
- ✓ Everything in Pro
- ✓ Dependency monitoring — watch vendor status pages (Stripe, AWS, OpenAI…)
- ✓ Up to 500 monitors
- ✓ Global checks — up to 5 monitors × 25 probe locations
- ✓ Per-location uptime + latency on status pages
- ✓ 25 status pages
- ✓ 15 team members — invite teammates, roles for owners, admins & members
Free plan: no card, forever. Paid plans start with a 14-day free trial — cancel anytime before it ends and you are not charged.
Questions, answered
We check on a schedule you control — as often as every 60 seconds on Pro, every few minutes on Free — from outside your network. As soon as a scheduled check fails or an expected heartbeat goes missing, we open an incident and alert your channels (after any grace you set).
Not for the Free plan — it is free forever with no card, so you can evaluate everything on it as long as you like. Every paid plan (Pro and Business) starts with a 14-day free trial: Stripe collects a card at checkout, but you keep full access for 14 days and are only charged if you do not cancel before the trial ends.
One thing we watch. From the outside: a website or API endpoint (HTTP status, body & response time), an SSL certificate, a domain or DNS record, an HTTPS redirect, or a TCP port. Or an inbound heartbeat your cron, queue, backup or app sends us — by curling a URL, or by emailing a unique address (so we can confirm your email is actually going out).
Alerts go to email (on by default), Slack, Discord, or any webhook you point us at — you can wire up several, and route them by tag. SMS, phone calls, and on-call schedules are on the roadmap, not available yet, so we do not promise them.
Yes — a branded public status page, on your own domain on Pro, with per-component uptime that updates automatically as incidents open and resolve. On Pro and Business, you can optionally let visitors subscribe to be emailed when something goes down or recovers.
Yes, on the Business plan. Add the vendors you rely on — Stripe, AWS, GitHub, OpenAI, your database host, plus 20+ more we already know how to read, or any public status-page URL — and Holter watches their status pages every few minutes. When one reports degradation or an outage (or recovers), your team is alerted on the same channels as your monitors, and you can show those upstream services on your own status page. Holter reads each vendor's published status page; it does not probe their systems directly.
Yes. Holter runs an MCP server — connect Claude Code, Cursor, or any MCP-capable agent with a team-scoped token, and your agent reads your repo locally and proposes the monitors (crons, queues, health checks, domains) plus the heartbeat code for you to review. Holter never sees your source — it only receives the monitors your agent sends, and you decide what to apply.
Stop finding out last.
Set up your first monitor in 3 minutes. Free for 5 monitors, no card required.
Start monitoring free