Automate Datadog
Turn Datadog's monitors, SLOs and usage figures into short reports the team will actually read. Monitors stuck in Alert or Warn, monitors someone muted and forgot, spikes in log ingestion and weekly error budget status arrive by email or chat. Let your AI assistant know what to report, check the plan before it starts, and it keeps reporting on its own.
Try asking
Create a ClickRuntime automation: every day at 8am, email me the Datadog monitors that are currently in Alert or Warn, with the monitor name, status and the service it covers.
Paste it into your assistant. It tells you what it will send or change before it runs live.
Ideas
What you can automate in Datadog
Get a morning list of firing Datadog monitors
Every day you see which monitors are in Alert or Warn, without opening Datadog.
See how it worksGet a weekly reminder of muted monitors
Each Monday you see monitors that are muted or under a downtime, so nothing stays silenced by accident.
See how it worksGet alerted when Datadog usage jumps
If log or host usage climbs well above normal, you hear about it the same day.
See how it worksGet a weekly SLO and error budget report
Every week you see each SLO's status and how much error budget is left, with the ones below target flagged.
See how it works Your own logicAny other automation you can describe in words
Your own rules, schedule and AI steps with Datadog, plus any other app. Describe it and your assistant builds it.
See how to build your ownBuild your own
Build any Datadog automation you can describe
Your own rules
Conditions, thresholds, filters and exceptions, written the way you'd explain them to a colleague.
Any number of apps
One automation can read from one app and update several others. Over 248 apps are ready, and anything with an API works too.
AI where it helps
Summaries, classification, extraction and drafts use ClickRuntime's built-in AI, with no extra key.
On a schedule or on each event
Runs as often as every 5 minutes, or the moment an app sends an event, and keeps track of what it already handled.
For example, ask your assistant
Every morning, check Datadog monitors that triggered overnight and send me a summary ranked by how long each was in alert.
When a Datadog monitor goes into alert, post the monitor name, value and graph link to Slack.
When a Datadog monitor stays in alert for 15 minutes, open a PagerDuty incident, create a Jira ticket, and post both links to our incidents Slack channel.
Beyond chat
Have Datadog work when you're not in the chat
On a schedule
Get a morning list of firing Datadog monitorsDaily
On a schedule
Get a weekly reminder of muted monitorsWeekly
On a schedule
Get alerted when Datadog usage jumpsDaily
Setup
Connect Datadog
Datadog needs two keys from your organization: an API key and an application key. Making the Datadog application key owned by a service account, with limited scopes, keeps access tight and independent of any one person.
Datadog
Developer / Observability
- In Datadog, open Organization Settings, then API Keys, and create a new key.
- Go to Organization Settings, then Application Keys, and create a new key, ideally owned by a service account.
- Optionally edit the Datadog application key's scopes to what the automations need, such as
monitors_read,slos_readandusage_read. - In the ClickRuntime dashboard, save both keys as secrets on Connections & secrets and choose the Datadog site your organization uses, for example US1 or EU.
Good to know
Datadog limits that affect automations
- Datadog runs separate sites (US1, US3, US5, EU, AP1 and US1-FED), and keys only work on their own site, so the connection must point at the right one.
- Reading data needs the Datadog application key as well as the API key.
- Some endpoints, such as Incidents v2, refuse application keys that have scopes set, so a scoped key may not cover every use.
- Usage and cost data may need a key with admin-level permissions.
- Rate limits differ by endpoint, so very frequent checks against a single endpoint can be throttled.
FAQ
Questions about automating Datadog
Why does Datadog need two keys?
Can it see monitors in real time instead of once a day?
What does ClickRuntime read from Datadog?
How is a usage spike detected?
Do I need an AI key for Datadog reports?
Is this a Datadog connector for ChatGPT or Claude?
Related pages
Keep exploring