Paging incident.io from GitHub

A lot of teams rely on CI/CD to catch issues early, but when something fails, that signal often dies quietly in the logs. I wanted a way to surface those failures where the rest of our incidents live, in the (wonderful) incident.io incident/response management tool.

I built incident.io-alert-action, a GitHub Action that sends alerts from your workflows directly into incident.io for routing and triaging.

It’s useful for things like:

• Deployments that fail halfway through

• Security or policy checks that block a release

• Long-running workflows that time out unexpectedly

• Health checks that you control via Workflows

Setup is one step:

- name: Send alert with custom data
  uses: lowlydba/incident-io-action@v1
  with:
    incident-io-token: ${{ secrets.INCIDENT_IO_TOKEN }}
    alert-source-config-id: ${{ secrets.INCIDENT_IO_ALERT_SOURCE_ID }}
    title: ‘High Error Rate Detected’
    status: ‘firing’
    description: ‘Error rate exceeded threshold’
    metadata: |
      {
        “service”: “api”,
        “environment”: “production”,
        “error_rate”: “5.2%”,
        “threshold”: “1.0%”,
        “region”: “us-east-1”
      }

Now those invisible failures become visible incidents, right alongside everything else your team responds to. Add whatever metadata you need to route the alert to the correct team, along with the context they need to effectively troubleshoot.

As always, sharing it as open source felt like the simplest way to make these signals visible and hopefully save someone else the same headache.❤️

It’s available here for use on the GitHub Action Marketplace:
🚀github.com/marketplace/actions/incident-io-alert-action

Share