Skip to content

Incident triage

The first fifteen minutes of an incident; an agent gathers context and on-call confirms severity and opens comms.

Source: scripts/dev/review-runbooks/incident-triage.yaml. CI checks it with runbook spec validate; teams, people, skills and called runbooks it names resolve against your server when you publish it.

apiVersion: therunbook.app/v1
kind: Runbook
metadata:
id: incident-triage
title: Incident triage
description: The first fifteen minutes of an incident; an agent gathers context and on-call confirms severity and opens comms.
owners: [team:oncall]
tags: [incident, oncall]
spec:
execution:
title: "Triage: ${{ inputs.summary }}"
inputs:
summary: { type: string, required: true }
service: { type: enum, options: [api, web, worker, billing], required: true }
defaults: { actor: agent }
steps:
- id: gather
title: Gather alerts, recent deploys and error spikes
instructions: |
List the alerts firing for ${{ inputs.service }}, the deploys of the last two hours and the top
error signatures since the first alert. Name the most likely suspect deploy, if any.
outputs:
alerts: { type: integer, required: true }
suspect_deploy: { type: string }
- id: confirm_severity
title: Confirm severity and incident commander
actor: human
assignee: team:oncall
instructions: Read the agent's summary, set the severity and name the incident commander.
checklist: [Severity set, Incident commander named, Customer impact described]
- id: comms
title: Draft the status page update
instructions: Draft a status page update from the confirmed severity and impact; keep it under 60 words.

Publish it to a project:

Terminal window
runbook push incident-triage.yaml --project <project> --draft
runbook publish incident-triage --project <project>