aegis/sre

Hackathon · Distributed Systems Track

Autonomous SRE Orchestration Swarm

Live incident console — trigger rollbacks, approve compensations, and watch verification complete in real time.

← Home connecting…

What you are looking at

Each row is one Temporal workflow for an incident: telemetry pull, AI/heuristic root-cause analysis, a drafted saga rollback plan, optional human approval, then compensating HTTP calls and a final verification pass.

Click Trigger incident to run the demo path end-to-end, or read the full story on the overview page.

  1. detectingPull trace spans for the alert via MCP.
  2. analyzingInfer root cause and draft refund/cancel steps.
  3. remediatingPlan is visible — Approve runs rollback; Ask re-runs analysis with your question (then back here).
  4. compensatingRun rollback activities; ✓/✗ per step.
  5. resolvedMock services and telemetry checked after rollback. Use Explain for an AI postmortem.

01 · Active incidents

workflow stream SSE · 3s poll
Incident ID Alert & trace errors Status Root cause Resolution Compensations Actions
No incidents yet — click Trigger incident to start one.