holmesgpt-catalyst

Diagnose auth-service pods entering CrashLoopBackOff, OOMKilled events, repeated pod restarts, and 401 authentication errors cascading across dependent services

Diagnose checkout-api pods that never start — ImagePullBackOff / ErrImagePull, "image not found" / "manifest unknown", pods stuck Pending or PodInitializing, a checkout-api deployment that won't roll out after a release

Diagnose api-gateway p99 latency spikes, SLO breaches, upstream timeouts, and HTTP 504 errors on the api-gateway service in the production Kubernetes cluster