TraceTriage

TraceTriage is a web app (optional Slack/Teams add-on) that sits on top of your existing monitoring stack and converts alert floods into a single, readable incident narrative. It ingests alerts, logs, traces, and deploy events from common tools, then groups them by likely shared cause, impact scope, and service ownership. Instead of another dashboard with 200 charts, it produces an “incident card” with: what changed, what broke first, blast radius, and the top three suspected causes with supporting evidence links. It also tracks SLO impact in plain language (“checkout error budget burned 18% in 30 minutes”) and suggests the next best queries/runbook steps. This is an AI + traditional app: AI for clustering/summarization and traditional rules for reliability and auditability. The goal is faster triage, fewer false escalations, and better postmortems with minimal new instrumentation.