Kubernetes Skills

**Judgement:** This skill provides a solid, fairly comprehensive toolkit for Kubernetes autoscaling that covers HPA, VPA, and KEDA with practical examples and troubleshooting guidance. It is well-suited for users who already understand Kubernetes concepts and want structured, tool-assisted autoscaling workflows. **Recommended scenarios:** - Teams running microservices on Kubernetes that need to implement or refine autoscaling strategies (CPU-based, custom metrics, or event-driven). - Cost-conscious environments where scale-to-zero, VPA right-sizing, and scheduled scaling can materially reduce cluster spend. - Operators/SREs managing multi-environment or multi-cluster setups who want consistent autoscaling patterns across staging/production. - Troubleshooting sessions where HPA/KEDA/VPA are not behaving as expected and quick inspection tools are valuable. **Key risks and limitations:** - **Operational risk:** Misconfigured HPA/VPA/KEDA (e.g., overly aggressive maxReplicas, incorrect thresholds, scale-to-zero on latency-sensitive services) can cause outages or severe performance degradation. - **Dependency on cluster components:** Requires metrics-server, VPA, and KEDA installations and CRDs; behavior will fail or degrade if these are missing or version-incompatible. - **Access and security:** Tools that apply or modify autoscaling resources need appropriate Kubernetes RBAC; misuse at high privilege levels can impact many workloads at once. - **Guardrails:** The documentation emphasizes patterns and best practices but does not show strong built-in safety guardrails (e.g., enforced replica caps, mandatory dry-runs), so it should be used by operators who understand the impact of autoscaling changes. Used with proper governance and review, this skill is a strong fit for production-oriented Kubernetes teams seeking structured autoscaling and cost-optimization capabilities.