sre-agent
claudecodexAgentic SRE orchestrator: evidence-first incident investigation across Kubernetes, Prometheus metrics, Loki or Elasticsearch/OpenSearch logs, Tempo/Jaeger traces, service-mesh state, and recent changes; ranked root-cause hypotheses; approval-gated remediation with rollback plans; validation, optionally under k6 load; and final incident reports.
1 skill · 1 command · 7 agents · 1 MCP server · hooks