Incidents come from where you already track them.
Alerts and tickets open incidents that wait for their fix. Every delivery is verified by signature or token before anything is stored.
Alert and ticket sources
- SentryNew error issues open incidents in Awaiting fix.Signed deliveries
- PagerDutyTriggered and resolved incidents open incidents here.Signed deliveries
- incident.ioLabsDeclared incidents open incidents here, signed with Standard Webhooks.Signed deliveries
- LinearLabsIssues labelled “incident” open incidents here.Signed deliveries
- RootlyLabsCreated and resolved incidents open incidents here.Token-authenticated
- OpsgenieLabsCreated and closed alerts open incidents here. Moving off Opsgenie? Your history still counts.Token-authenticated
- DatadogLabsTriggered monitors open incidents here.Token-authenticated
- JiraLabsBugs and incidents created in Jira open incidents here.Token-authenticated
- Generic webhookLabsAnything that can send a signed JSON webhook (Standard Webhooks).Signed deliveries
And around the loop
- GitHubIssues become incidents, pull requests get one check and one comment, proven tests arrive as pull requests.
- Slack and signed webhooksProofs, blocked pull requests, failing tests and a weekly digest, sent where your team already looks.
- Claude Code, Cursor and CopilotIncident memory for coding agents over MCP, and lessons synced into AGENTS.md.
- GitHub ActionsRun pull request checks on your own runners with the Aftershock action.