Triage a PagerDuty Alert in Minutes
For noisy alerts that might be real. DuckyHelper acknowledges it, reads the service graphs in Datadog, checks traffic and deploys, and tells you whether it's a blip or a problem, with the runbook step to try first.
“Triage this PagerDuty alert: check Datadog against traffic and deploys and tell me if it's real.”
- 01Turn PagerDuty and Datadog into APIs it can callIt watches each site's own requests once and saves them as actions. Kept on your device.
acknowledge(incident)getIncident(id)queryMetrics(query, range) - 02Acknowledge the alert
- 03Read the service graphs
- 04Check traffic and deploys for the same time
- 05Say blip or problem, with a reason
- 06Draft the runbook step and the #incidents postPosting waits for your OK.
You getKnow in minutes. Real alerts get handled fast and noise gets closed. The next alert is instant.
A run, drawn from this page · early access on Mac
It's built to wait for your OK.
It does the work, then stops with the Slack post ready for you. You see exactly what it wrote, then Send, Edit or Deny.
If it hits a sign-in, you sign in yourself. It's built not to type your passwords or card numbers.
Do it once. Then let it repeat.
When it's done, it offers to keep doing it: whenever a new one comes in. Each run goes through the API it learned the first time, so it's quick, and it's still built to ask before it sends or pays.
Always-on automations are in early access on Mac.
Before you ask it.
What do I say?
Press fn (Ctrl+Shift+Space on Windows and Linux) and ask out loud, in your own words, like "Triage this PagerDuty alert: check Datadog against traffic and deploys and tell me if it's real."
Why is the second time faster?
The first time, it writes small JavaScript tools for PagerDuty, Datadog and Slack and builds an API from each site's own buttons, forms and requests. After that it calls them directly, so the job runs in one go instead of click by click. It costs less and breaks less.
Will it send anything without me?
It's built to ask first before it sends a message or pays for anything, and to wait for your yes. It may not recognize every send or payment, so keep an eye on tasks that involve messages or money. It's built not to type your passwords or card numbers: you sign in yourself, and it picks up from there.
Why not just do it by hand?
Most alerts are noise, but you can't tell until you've opened the graphs and checked what changed.
Can it keep doing this on its own?
Yes. When it's done it offers to make it automatic (whenever a new one comes in). Each run goes the same way and is still built to ask before it sends or pays.
Can I use this today?
Doing jobs in your apps is in early access on Mac. Talking to DuckyHelper, and having it see your screen and draw on it, works today on Mac, Windows and Linux.
- Early access
“Customers report checkout errors. Check Datadog, Intercom and recent deploys, and draft a Slack update.”
Investigate an Incident Across Every Dashboard
See it run - Early access
“Acknowledge this PagerDuty page, pull the Datadog dashboard and recent deploys, and draft a war-room post.”
Answer an On-Call Page With a Head Start
See it run - Early access
“Set up access for our new designer: an Okta user with design groups, the Bitwarden design collection, and Slack channels.”
Set Up a New Employee's Access
See it run
Do anything
Ask it a question or hand it a job, in any app.
Doing jobs for you is in early access on Mac. Asking works on Mac, Windows and Linux.
