Watch the walls
so they don't.
Merlonix is autonomous monitoring for the infrastructure your agency forgot to monitor. SSL expiry, DNS drift, and upstream vendor outages — one pane of glass, every client, one catch it before they do alert flow.
No credit card for the trial. Cancel any time.
- Time to first alert
- 5 min
- Independent resolvers
- 3
- Upstream vendors watched
- 11
- Gross margin
- 94.9%
The gaps between other monitors
You already have uptime monitoring. You still get surprised.
The things that break your clients don't look like the things uptime monitors catch. A 200 OK can sit on top of an expired cert for hours. A DNS record can drift with zero HTTP signal. A vendor outage can tank half your stack while your ping-every-60s dashboard stays green.
Expired silently
SSL certificates
Let's Encrypt renewals fail. Manually-installed wildcards lapse. Chains break after an upstream CA rotation. You find out when a client's checkout stops working.
Drifted quietly
DNS records
Someone ran a migration. Someone deleted an A record. Someone flipped a CNAME. Now a subdomain resolves to nothing and nobody at the agency knew until the client called.
Degraded upstream
Vendor status
Stripe is having a partial outage. Cloudflare's DNS in one region is flapping. Shopify's admin is intermittent. Your team learns on Twitter.
How it works
Four moving parts. No dashboards to check.
Merlonix is designed around only alert when it matters. Everything else is just the plumbing that lets us say that with a straight face.
01
Attest once per asset
You tell Merlonix which domains you're authorized to watch — inline text, DNS TXT, or an .well-known file. The attestation is signed and stored. This keeps us on the right side of the line between monitoring and snooping.
02
Agents check every 5 minutes
Cloudflare Workers fan out to three independent DNS resolvers, pull each hostname's leaf cert, and diff against the expected DNS records. Eleven upstream vendor status pages are polled globally on one shared schedule — you only pay the fetch once per vendor.
03
AI triages before we alert
An LLM classifier decides: real issue, false positive, informational, or uncertain. False positives go to a digest. Real issues page you. Uncertain ones get a second opinion from a different model. You only see the ones that need you.
04
Deliver to where you already are
Email, Slack, webhook, or SMS — whichever channel the client-facing engineer actually reads. Ack / mute / resolve from the alert itself. No "please log in to the dashboard to triage."
Pricing
One flat fee per month. Every client included.
No per-seat upsells. No per-check metering. Pick the tier that fits your asset count and move on.
Starter
For solo operators watching a handful of clients.
$29/ month
- 10 monitored assets
- 1 seat
- 15-min check cadence
- Email + webhook alerts
Team
For small agencies that need faster checks and more seats.
$79/ month
- 50 monitored assets
- 5 seats
- 10-min check cadence
- Slack + webhook alerts
- API access
Agency
For the agencies that can't afford an outage they didn't see.
$199/ month
- 200 monitored assets
- 15 seats
- 5-min check cadence
- All alert channels
- API access
- Priority support
Start watching for free.
Pick a tier. Add your first client domain. Get the first alert within 10 minutes. Cancel any time during the 14-day trial and pay nothing.