Website Care · Starter

Get alerted when a site breaks

You want to hear the site is down (or a certificate or domain is expiring) before a customer tells you.

  • Any
  • Claude Code
  • Grok Build
  • Cursor
  • Codex CLI

For: Owner, Office manager, Tech person, Anyone at work · Updated November 10, 2026

A starting point: fill in the [brackets] with your own details, and check the answer before you use it. Tool screens change often, so yours may look a little different.

The prompts

Copy a quick prompt, or grab the full skill below.

QUICK PROMPT 1
Why this prompt works: Defining 'broken' first gives the assistant clear criteria.
For the site I work on, list what 'broken' would mean (home page down, form not sending, and so on) and suggest one simple check for each. Plan only; don't sign up for anything.
QUICK PROMPT 2
Why this prompt works: Setting a budget limit keeps suggestions realistic.
Suggest a free or low-cost way to get alerted when the site I work on goes down or its security certificate or domain is about to expire. Explain the options plainly before setting anything up.
QUICK PROMPT 3
Why this prompt works: Asking for a draft keeps you in charge of the final decision.
Who should get alerts for the site I work on, and how, so they aren't ignored? Draft a short alert plan with a main and backup contact for me to review.

What this covers: A lean alert plan: uptime and keyword checks, key forms and flows, certificate and domain expiry, scheduled-job heartbeats, error tracking, and who gets each alert.

For anyone improving a site or app, whether a small-business owner or someone at a larger company practicing better prompts. Without monitoring, the first person to notice the site is down is often a frustrated customer. Paste a prompt into any AI assistant that can see the site or repo.

Two ways to use it

  1. Any AI chat (Copilot, ChatGPT, Claude, Grok): copy one of the quick prompts above. Each one asks for a plain-language report first, so nothing changes until you say so.
  2. A coding agent working in the site’s code (Claude Code, Grok Build, Cursor, Codex CLI): save the full skill below as SKILL.md in a folder named uptime-and-monitoring-setup, for example .claude/skills/uptime-and-monitoring-setup/SKILL.md (Grok Build also reads that location, or use .grok/skills/). Then ask the agent to use the uptime-and-monitoring-setup skill. It audits first, makes the smallest safe changes in a review branch, and ends with a report.

Download SKILL.md · All Website Care skills

The full skill (SKILL.md)

Last reviewed Nov 10, 2026. Standards and tools change, so check anything dated before relying on it.

---
name: uptime-and-monitoring-setup
description: Use when asked how the owner will know a website or web app is down, broken, slow, or expiring - uptime and keyword checks, form and key-flow checks, TLS certificate and domain expiry, error tracking, Core Web Vitals field monitoring, build/deploy failure alerts, and who gets alerted. Designs a lean monitoring plan using free tiers where possible and proposes config as a review PR. Does not sign up for services or spend money without approval.
---

# Uptime and monitoring setup

Last reviewed: 2026-11-10

## Quick prompts

For anyone improving a site or app, whether a small-business owner or someone at a larger company practicing better prompts. Without monitoring, the first person to notice the site is down is often a frustrated customer. Paste a prompt into any AI assistant that can see the site or repo.

1. "For the site I work on, list what 'broken' would mean (home page down, form not sending, and so on) and suggest one simple check for each. Plan only; don't sign up for anything."
   *Why this prompt works:* Defining 'broken' first gives the assistant clear criteria.

2. "Suggest a free or low-cost way to get alerted when the site I work on goes down or its security certificate or domain is about to expire. Explain the options plainly before setting anything up."
   *Why this prompt works:* Setting a budget limit keeps suggestions realistic.

3. "Who should get alerts for the site I work on, and how, so they aren't ignored? Draft a short alert plan with a main and backup contact for me to review."
   *Why this prompt works:* Asking for a draft keeps you in charge of the final decision.

## Hard limits

- Do not create accounts, enter payment details, or add the owner's phone/email to a service without explicit approval.
- Synthetic checks must not submit real forms repeatedly or create junk records; use a test mode, a health endpoint, or a dedicated test inbox the owner approved.
- Keep alerts few and actionable; an ignored alert is no alert. Never put secrets in monitor URLs.

## Inputs to gather

1. What "broken" means for this site: home down, a key page down, search broken, forms not delivering, checkout failing, login failing.
2. Who should be alerted, how (email, chat, SMS), and quiet hours.
3. Hosting/CDN features already available (analytics, health checks, notifications on failed builds).
4. Domain registrar and renewal settings; certificate issuer (usually automatic).

## Interface-type notes

- **Content sites**: uptime on home plus one article; keyword check that the page body contains expected text (catches blank or error pages that return 200); scheduled-build success; sitemap reachable.
- **Marketing**: landing pages and form delivery; tag manager failures do not take the page down but break attribution.
- **E-commerce**: checkout synthetic in test mode, payment provider status page, inventory sync jobs.
- **SaaS / app**: login synthetic with a test account, API health endpoint, background job queues, error rate from error tracking.
- **SPA**: a 200 shell can hide a broken app; check for a rendered element with a browser-based check.

## Steps

1. **Map critical paths** and pick one check for each (HTTP status, keyword, browser step, or heartbeat for scheduled jobs).
2. **Choose tools**: host's built-in notifications, a free uptime monitor, GitHub Actions scheduled workflow for simple checks, error tracking with a free tier, real-user Web Vitals via the `web-vitals` library into existing analytics.
3. **Expiry**: TLS certificate expiry alert (even with auto-renewal), domain expiry and auto-renew on, payment card on the registrar valid (owner checks).
4. **Scheduled jobs**: heartbeat ("dead man's switch") monitors for cron builds and backups so silence triggers an alert.
5. **Alert routing**: one primary contact, one backup, a short note in each alert saying what to check first (link to the runbook).
6. **Status and runbook**: optional public status page; a one-page "site is down" runbook linking to backup-and-rollback-readiness.

## Validation

- Trigger each alert once in a safe way (monitor a test URL that returns 500, pause a heartbeat) and confirm delivery to the right person.
- Confirm checks do not count toward rate limits or analytics in a misleading way (exclude monitor user agents from analytics if possible).

## Report format

1. Critical paths and the check for each (type, URL/endpoint, interval, expected result).
2. Tools proposed with cost (free tier limits) and what needs owner sign-up.
3. Alert routing table. 4. Config as a review PR (for example a scheduled workflow), "not merged". 5. Test results of each alert.

## References

- Google SRE book, Monitoring Distributed Systems: https://sre.google/sre-book/monitoring-distributed-systems/
- web-vitals library: https://github.com/GoogleChrome/web-vitals
- GitHub Actions scheduled workflows: https://docs.github.com/en/actions/writing-workflows/choosing-when-your-workflow-runs/events-that-trigger-workflows#schedule
- Let's Encrypt expiration emails ending, so monitor yourself: https://letsencrypt.org/2025/01/22/ending-expiration-emails/
- ICANN domain renewal basics: https://www.icann.org/resources/pages/domain-name-renewal-expiration-faqs-2018-12-07-en

Check before you send

  • It gave a plain-language report before suggesting or making any change
  • Each finding names the exact page or file and why it matters
  • It said what it could not check or verify
  • Nothing was changed, published, or deleted without your OK
  • Fixes are small and keep the existing design and content

Privacy note: what not to paste

Paste page addresses, screenshots, or reports, not passwords, API keys, customer data, or private admin links. Only check sites you or your employer control, and get the site owner’s OK before anyone changes anything.