Cron monitoring · checked 2026-10-14

Can I vibecode Cronitor?

YES· one-shottable

A cron monitor is a ping receiver with a clock: each job calls a URL when it finishes, and a loop raises an alert when the next call doesn't arrive in time. An agent builds that, with cron-expression schedules, start and fail pings and email, Slack or ntfy alerts, in one sitting. The real requirement is where it runs: on a machine that isn't the one hosting the jobs, and ideally watched by a free hosted heartbeat itself. Healthchecks, which has a hosted free tier of 20 jobs and a BSD-licensed server, may save you the build altogether.

$15/moprice

$180you'd save a year

one sittingbuild time

—people replaced it

Before you start

What you need on hand.

DifficultyIntermediate · Some system setup or keys.

TimeAn hour or two for the first version

Claude Code, Codex or CursorAny of them can run the prompt.
Node.js 22better-sqlite3 compiles a native module, so build tools must be installed.
A small always-on serverA VPS, or a home machine reachable by your jobs; not the host the jobs run on.
An alert channelSMTP, a Slack webhook or a free ntfy topic.
Your list of jobs and schedulesExport it from Cronitor as YAML.

Before you buildCronitor Hacker already covers 5 monitors for free, and Healthchecks.io covers 20. If your job list is that short, you can skip the build.

1. Install one agent: Claude Code, Codex, Cursor.
2. Make an empty folder and open it in the agent.
3. Paste the prompt below, press enter, and answer its questions as it builds.

The prompt

Build me a cron job monitor to replace Cronitor. Requirements:

- Node.js 22 + TypeScript: a Fastify server with server-rendered pages and no front-end
  framework. `npm start` serves on http://localhost:4750.
- Storage: better-sqlite3 in ./data/monitor.db: monitors (name, token, schedule, grace_minutes,
  status, paused), pings (monitor, ts, kind, exit_code, duration_ms, message, source_ip_hash) and
  alerts sent. Pings older than 90 days are deleted nightly.
- Schedules: "every N minutes or hours" or a cron expression parsed with `croner`, evaluated
  in a per-monitor IANA timezone. Each monitor has a grace period in minutes.
- Ping endpoints, any HTTP method: /ping/<token> (success), /ping/<token>/start,
  /ping/<token>/fail and /ping/<token>/<exit-code>. Store up to 10 KB of body as the message and
  compute duration from start to success. Unknown tokens get a plain 404.
- Status: new, up, late (past schedule), down (past schedule plus grace), failed (fail ping or a
  start with no finish within MAX_RUNTIME). A checker runs every 15 seconds and sends one alert
  when a monitor goes down and one when it recovers, never repeated while the state is unchanged.
- Alerts, per monitor: email (nodemailer SMTP from .env), Slack-compatible webhook, ntfy topic and
  generic JSON webhook. Pause switch and maintenance windows.
- Pages: monitor list with status dots, a detail page with the last 100 pings, a 90-day uptime bar,
  and a copy-ready snippet like `curl -fsS -m 10 --retry 5 https://host/ping/<token>`.
- Admin: one password in .env (signed cookie). `GET /healthz`, and an optional outbound ping to
  HEARTBEAT_URL every 5 minutes so you hear if the monitor itself dies.
- Single user. No accounts, no telemetry, local-first. Out of scope: website checks, status pages,
  SMS and on-call escalation, teams.
- README: a crontab example, a systemd unit, backups, and why it must run on a machine other than
  the jobs it watches.

Opens your agent with the prompt filled in, you press enter. Or copy it and paste it into an empty folder.

Share on X

Max MiskowSponsoredAI agents, automations and MVPs, built for your company.Visit site ›

Then keep going

Next prompts, same folder.

The first prompt builds the core. Run these one at a time in the same folder to grow it into your own Cronitor.

Add a Cronitor importer to the monitor in this folder. Requirements:

- `monitor import-cronitor monitors.yaml` reads the YAML from Cronitor's monitors API
  (GET /api/monitors?format=yaml) and creates one monitor per entry.
- Map name, key, schedule, timezone and grace seconds; skip types that aren't jobs or
  heartbeats and list them.
- Print the new ping URL for each monitor so I can swap it into the crontab, and write the
  mapping to ./data/import-map.csv.
- Dry-run by default; --yes writes.
- Update the README.

Add a small wrapper CLI to the monitor in this folder. Requirements:

- `ping-run <token> -- <command...>` calls /start, runs the command, then calls success or
  /fail with the exit code and the last 2 KB of stderr as the message.
- Take MONITOR_URL from the environment, retry pings 3 times with backoff, and never let
  a failed ping change the command's own exit code.
- Ship it as a single file that needs only Node.js, plus a shell version using curl.
- Update the README with crontab examples.

Add a weekly summary to the monitor in this folder. Requirements:

- Every Monday 08:00 in TIMEZONE from .env, send one plain-text email with, per monitor, the
  number of runs, late runs, failures and the slowest duration.
- Highlight jobs that never pinged during the week.
- A `monitor summary --now` command for preview.
- Update the README.

What you lose

SMS and on-call alerts through PagerDuty, Opsgenie and Splunk On-Call
Hosted status pages with subscribers
Website, API and synthetic browser checks at 30-second intervals
Dashboard roles, SCIM and SAML sign-in for teams
A service that stays up when your own server doesn't

Why people still pay

A monitor on your own box can die with the box. Cronitor is somebody else's uptime, with SMS and on-call tools that wake a person when everything else is down, and teams add the status pages and user roles.

Moatinfrastructure scale integrations

Do the math

Cronitor or your own copy?

Over

Your usage

Count my time3h ×$/h

Cronitor, 1 yr$180

Your own copy, 1 yr$41running + build session

You keep$139pays for itself in the first month

Estimate · A small VPS is about $5 a month, and nothing extra if you already run one. Alerts through SMTP, Slack or ntfy are free at this volume. The build is token usage on your agent plan.

Leaving Cronitor

Take your data, then cancel.

1 · Take your data out of Cronitor

1Take your API key from your Cronitor account.
2Run `curl -X GET "https://cronitor.io/api/monitors?format=yaml" -u API_KEY:` (or send an Accept: application/yaml header) to download every monitor as YAML.
3Each monitor has a key and one of four types: job, check, heartbeat or site. Cronitor's ping URL looks like https://cronitor.link/ping/API_KEY/<monitor key>.
4The documented export covers monitor definitions; ping history, alert history and status page content aren't mentioned, so screenshot anything you need.

2 · Cancel Cronitor

1Cancel the renewal from your online account management page, or by contacting Cronitor support.
2Do it before the renewal date: you keep access through the end of the billing cycle, and paid fees are non-refundable except where the law says otherwise (support may still decide case by case).
3The pricing page says there are no commitments and you can switch plans or cancel anytime; Hacker is the free plan with 5 monitors.

Change the ping URL in every crontab, CI job and script before you cancel: after that, a job that pings Cronitor reports to nothing.
Run the old and the new monitor in parallel for at least one full cycle of your slowest job, so a weekly job gets a week.
Cronitor's terms don't say what happens to your data after cancelling; ask support at support@cronitor.io if retention matters.
Trim the account to 5 monitors if you plan to stay on Hacker.

Checked 2026-10-14 against Cronitor's docs: 1, 2, 3, 4

The escape hatch

Don't feel like building it? These are already free.

HealthchecksOpen sourceBSD-3-Clause cron monitoring server in Django with 25+ integrations; the hosted free plan covers 20 jobs.★ 10k · active Oct 2026 · docker to self-host
WebVisit ↗Uptime KumaOpen sourceSelf-hosted monitor with push-style monitors for jobs that report in, plus 90+ notification services; MIT-licensed.★ 92k · active Oct 2026 · docker to self-host
WebVisit ↗GatusOpen sourceApache-2.0 status page and alerting tool whose external endpoints take pushed results and flag missing heartbeats.★ 12k · active Oct 2026 · docker to self-host
WebVisit ↗

The numbers

Cronitor pricing

Plan	Monthly	Annual, per month	What you get
Hacker	$0	$0	5 monitors, 1 dashboard user, email and Slack, Discord, Teams and webhook alerts, 1 basic status page with 50 subscribers, 5-minute check frequency.
Business	$15	—	$2 a month per monitor and $5 a month per user with no base fee, so 5 monitors and 1 user is $15. SMS and on-call alerts, 30-second checks, 12 months of retention, unlimited API requests.
Enterprise	—	$500	Custom integrations, a dedicated engineer, priority support, unlimited status pages.

Free tier: Five monitors, one user, 5-minute checks, email and chat alerts only.
Billing: Business is billed monthly on the monitors and users you actually have, with no contract. Enterprise is annual invoice billing. Subscriptions renew automatically.
Hidden costs: A branded status page is $25 a month per page and a private one $50. Synthetic browser checks are $1 per 1,000, web analytics beyond 100,000 events is $10 per 100,000, and every extra monitor or user adds to the bill.
Source: cronitor.io

Also on the list

Better Stack$34/mo · UptimeYES

Airtable$24/mo · DatabasesYES

Carrd$1.58/mo · Website buildersYES

Questions

Cronitor, briefly

Not answered here? Tell us and we'll add it.

Ask a question ›

Hacker covers 5 monitors and one user, with email and Slack alerts, a basic status page and 5-minute checks.

Yes. An AI coding agent such as Claude Code, Codex or Cursor can build a usable personal version in one session with the prompt on this page, running on your own machine.

About $15 a month on the Business (1 user, 5 monitors) plan, $180 a year (checked 2026-10-14).

SMS and on-call alerts through PagerDuty, Opsgenie and Splunk On-Call; Hosted status pages with subscribers; Website, API and synthetic browser checks at 30-second intervals; Dashboard roles, SCIM and SAML sign-in for teams; A service that stays up when your own server doesn't. If any of those matter to you, keep paying.