Pulsecheck watches your scheduled jobs, workflows and AI agents. If one doesn't run, or runs but produces rubbish, you hear about it in minutes, not days.
No card needed · Set up in two minutes
Most monitors only tell you a job didn't run. Pulsecheck also tells you when it ran and quietly did nothing useful.
Every job gets its own ping URL. If the ping doesn't arrive on schedule, plus a grace period you choose, we alert you.
Send a little JSON with the ping, like {"count": 12}. Set rules such as "at least 1 row" or "must include email". Break a rule and we alert, even though it "succeeded".
Call the /fail URL from your error branch and we flag it straight away, with whatever details you send.
n8n, Make, Zapier, GitHub Actions, cron, Claude Code and other agent routines, Python or Node scripts. Copy a snippet, paste it in, done.
From sign-up to your first job going green.
Alerts when something breaks, and again when it recovers.
Recent check-in history for every job, with the problems we found.
Prices in US dollars, billed monthly. Cancel any time.
For AI agents and serious workflows
For automation agencies
Just add one request at the end of the run. For output checks, include a small JSON summary of what the run produced.
Only what you send in the ping, plus the last 30 check-ins for each job. Keep it to counts and field names, not customer data.
Yes, any time. You drop back to the free plan and keep your first 3 jobs.
Set up your first monitor in two minutes. Free for 3 jobs.
Start free