Oh no! Something has gone wrong, and we need to respond! We can do that in a few ways:

1. Using /incident

2. From a message directly

3. From a workflow

You'll be asked for some basic information about the incident - all totally optional.

By default we'll kick off a low-severity incident with a randomly generated name, which is the quickest way to put all our tooling at your fingertips. Both the name and severity are easy to change later. We'll also post in the incident announcements channel (#incidents by default).

This announcement channel post links to the incident homepage and shows a number of useful pieces of information, which we will update as the incident unfolds.

<aside> 💡 If you're using /incident to kick off your incident, and you know what's wrong, you can type extra text at the end to pre-fill the form that pops up when you create an incident 💅🏻

For example, type /incident The website is down and you'll get something that looks like the image above.

</aside>

Running a test incident 🧪

Sometimes, you just want to practice with incident.io without the whole company knowing, and without messing up your lovely incident library with fake data.

Luckily, we support test incidents for exactly this use case! To get started, create an incident with /incident test.

Type /incident test to start a test incident.

Test incidents are almost exactly the same as normal incidents: you can set summaries, leads, escalate to others, and receive nudges. The only differences are

• We won’t announce it in the #incidents channel