HolmesGPT/holmesgpt: CNCF Sandbox SRE Agent for Kubernetes Root Cause Analysis — 2,113 Stars

GitHub·medium signal

HolmesGPT is an agentic troubleshooting tool for Kubernetes and cloud-native environments, accepted into the CNCF Sandbox in October 2025 with major contributions from Microsoft. Unlike static dashboards, it actively decides what observability data to fetch, runs targeted queries, and iteratively refines root-cause hypotheses. Works with any stack — K8s, VMs, cloud providers, databases. The CNCF acceptance legitimizes AI-for-ops as a production category, not just a demo.