Incident Triagev0.3.0

When a PagerDuty alert fires for high error rates in production, an engineer uses Incident Triage to classify the severity, scope the affected users, correlate with recent deploys via gh CLI, and investigate using runbook references to resolve the incident quickly.

A failed GitHub Actions deployment triggers a custom webhook. The skill parses the alert, checks recent code merges, and guides the on-call engineer through rollback or hotfix steps while communicating with the team.

AWS CloudWatch detects an unexpected spike in latency. The skill extracts alarm details, correlates with EventBridge events, and walks through log analysis and dashboard checks to identify the root cause.

Sentry reports a new error in a critical service. The engineer uses the triage framework to classify the error, determine blast radius, check for related changes, and escalate if necessary, all while ensuring runbook steps are followed.

A custom monitoring tool sends an alert about disk usage exceeding threshold. Incident Triage parses the webhook, applies the five-step process, and triggers escalation based on severity, integrating with existing on-call tools.

Offer a managed service where your team handles incident triage for clients using this skill. Clients send alerts via webhooks, and your team uses the structured framework to respond, create tickets, and escalate, ensuring SLAs are met.

Consult with companies to customize the escalation guide and runbook template for their infrastructure. Provide training on the triage process and integrate with their monitoring stack (PagerDuty, Datadog, etc.) for a one-time setup fee plus ongoing support.

Develop and sell a plugin for DevOps platforms like PagerDuty or Opsgenie that embeds Incident Triage as a guided workflow. Users pay per-action or per-seat to reduce MTTR and standardize incident response.