How is this different from Better Stack's built-in notifications?

Built-in alerts route to fixed channels. Here the agent decides the recipient and the wording per monitor, so a specific failure reaches a specific person, with copy you write, without standing up a separate escalation policy for each case.

Will a long outage spam the recipient?

Not if you track status. The agent emails on the down edge and again on recovery, and stays quiet in between, so one outage produces one alert and one all-clear.

How fast does the email go out?

Within your polling interval. The server answers the agent's call rather than pushing, so a one-minute loop means the email lands about a minute after a check starts failing.

Better StackResend

Send Resend emails when a Better Stack monitor goes down

A Better Stack + Resend agent flow

A monitor flips to down and the right person is the one not in the on-call chat: a customer's account manager, a vendor's NOC, an exec who wants a heads-up when checkout is unreachable. This recipe lets your agent read Better Stack's Uptime monitors and email the people who belong on a specific incident, with wording you control, rather than the generic alert everyone tunes out. It calls uptime_list_monitors, finds the ones currently down, and sends a message per affected monitor with Resend's send-email: which check failed, since when, and a link to the monitor. The Better Stack server replies when asked, not on a push, so the agent runs on a tight loop and the email goes out shortly after a check starts failing.

The flow

Better Stackuptime_list_monitors

Lists Uptime monitors, with filtering options.

Resendsend-email

Sends a single transactional email immediately or scheduled, with HTML/text, attachments, CC/BCC, and tags.

Step by step

Decide who hears about what
Map monitors, or groups of them, to recipients and a from address. The checkout monitor might email the revenue lead; the API monitor might email a partner. Set a verified sending domain in Resend first.
Read the monitors
The agent calls uptime_list_monitors and inspects each one's status, reading which are down, when they last changed, and the monitor's name and URL.
Send the email
For each newly-down monitor it calls send-email with the mapped recipients, a subject naming the check, and a body with the failure time and the monitor link. Use a tag so these are filterable in Resend.
Email on the edges, not every poll
Track each monitor's last-known status. Email when one transitions down, and again when it recovers, instead of resending while it stays down, so a long outage is two emails, not forty.

Tell your agent

Every minute, list Better Stack Uptime monitors and find any that just went from up to down. For each, send a Resend email from alerts@acme.com to the owner mapped for that monitor, subject 'DOWN: <monitor name>', body with the time it failed and the monitor link, tagged 'uptime'. Send a recovery email when it comes back up. Don't resend while it stays down.

Setup

This flow needs both servers connected to your agent. Follow each install guide:

Better Stack install guide Resend install guide

Worth knowing

send-email needs a verified sending domain in Resend with SPF and DKIM, or the mail won't deliver. Set the from address to that domain and confirm verification before relying on it.
uptime_list_monitors reports current status, not a stream of changes. The agent must compare against the status it saw last run to detect a real down or recovery edge.
Add a confirmation rule for who can be emailed. Pin the recipient mapping in the prompt so a misread monitor name can't send a real outage notice to the wrong customer.

Questions

How is this different from Better Stack's built-in notifications?: Built-in alerts route to fixed channels. Here the agent decides the recipient and the wording per monitor, so a specific failure reaches a specific person, with copy you write, without standing up a separate escalation policy for each case.
Will a long outage spam the recipient?: Not if you track status. The agent emails on the down edge and again on recovery, and stays quiet in between, so one outage produces one alert and one all-clear.
How fast does the email go out?: Within your polling interval. The server answers the agent's call rather than pushing, so a one-minute loop means the email lands about a minute after a check starts failing.