Introducing Pingstack

The failures that hurt most are the quiet ones. A nightly backup stops running weeks before anyone needs it. A certificate renewal fails behind a CDN. A server gets rebuilt and nobody puts the crontab back. Your first alert is a customer.

Jacob and I have run SaaS apps for over 20 years, and we watched them the hard way. We had one service for uptime, another for cron jobs, a dead man’s switch for the scripts nobody remembered, and a calendar reminder for every certificate and domain renewal. Each one had its own alerts, its own bill and its own idea of who should get woken up. Things still slipped through the cracks. Keeping it all wired together was a pain in the ass.

So we built the one tool we actually wanted. It watches the jobs and services you can’t afford to forget.

What it does

Heartbeat checks for cron jobs and scheduled tasks. Each job gets a private ping URL, and the job calls it when it finishes. If that call doesn’t show up on schedule (an interval, or a cron expression in a time zone you pick) plus a grace period, you get an alert. A job can also send a number with the ping, like a backup’s size or a queue’s length, and you’ll hear about it when that number crosses a threshold.

Active checks for everything else. Websites and APIs get checked from three probe regions: US East, EU West and AP South. TCP checks make sure a port accepts connections. DNS checks compare a record across the Cloudflare, Google and Quad9 resolvers. SSL certificate and domain checks look up expiry dates every hour and warn you at the number of days you pick.

Alerts that go to the right person. Checks alert a notification group, so the whole team doesn’t get pinged for everything. Groups deliver by email, Slack, Microsoft Teams, Discord and webhooks on every plan, through PagerDuty, Jira Service Management and Jira tickets on Team and Business, and by SMS and phone call on paid plans. Rosters and schedules make sure the person on call hears about it and everyone else gets to sleep.

Status pages and wallboards. You can share any dashboard publicly with your name, colors and logo, or put it on a TV as a wallboard.

Your agent can set it up

A lot of developers now ask an agent to write the job. In my experience the agent almost never adds monitoring unless you ask. Pingstack has a remote MCP server with browser sign-in, so Claude Code, GitHub Copilot and other MCP clients can create checks, add them to dashboards and read incidents, uptime and latency.

Connecting Claude Code takes one command:

claude mcp add --transport http pingstack https://app.pingstack.io/mcp

Then ask it to “add a heartbeat check for the nightly export job and a report step that reads the ping URL from a secret”. It stays inside your role’s permissions and your plan’s limits. It can’t delete checks, publish a status page or buy anything. When you hit a plan limit, it hands back an upgrade link for a person to open. The Claude Code recipe walks through it, and there are recipes for 25 schedulers, from crontab and systemd timers to Kubernetes, Laravel and GitHub Actions.

Pricing

Free covers 15 checks, with the API, MCP and every alert integration except SMS, calls, PagerDuty and Jira. Pro is $12 a month for 50 checks. Team is $29 a month for 250 checks and 5 seats. Business is $79 a month for 1,000 checks and 15 seats, then $5 a seat. Pay yearly and you get 2 months free. The full comparison is on the pricing page.

What it doesn’t do

Pingstack isn’t an observability suite. There’s no APM, log search or tracing, and active checks can’t reach hosts inside a private network yet. Heartbeats work anywhere with outbound HTTPS, so jobs behind a firewall are fine. If you need SSO or a signed SLA today, we’re not the right fit yet.

Try it and tell me what’s missing

Start free and add a heartbeat to the job you’d least like to fail silently. Is something confusing, broken or missing? Tell me. I read every message, and the changelog shows what changes because of them.