Datadog Alert Explainer with Linear Follow-Up

Combines instant alert explanation with durable follow-up. The bot explains why a Datadog monitor fired, identifies the likely deploy cause, and then turns that diagnosis into a tracked Linear issue so the underlying problem is not forgotten once the page clears.

Use it for alerts that need investigation beyond the immediate response, where teams routinely lose track of root-cause work after the fire is acked. Best for squads that triage in Slack and plan in Linear.

The operator asks the bot to explain an alert in Slack. The agent fetches the monitor state, breached query, and dashboard from Datadog, then pulls recent deploys and commits from GitHub to form a hypothesis. It posts the explanation in the thread. A logic step checks whether the alert looks like a real defect versus transient noise. If it is a likely defect, the agent creates a Linear issue prefilled with the monitor link, suspect deploys, and metric snapshot, and replies with the issue link for the responder to confirm.