Trigger: Use when working with Google Kubernetes Engine basics — cluster setup, node pools, workloads, networking, and storage fundamentals.
GKE Basics
Managed Kubernetes platform on Google Cloud. Defaults to Autopilot mode.
Quick Start
gcloud services enable container.googleapis.com --quiet
gcloud container clusters create-auto my-cluster --region=us-central1 --quiet
gcloud container clusters get-credentials my-cluster --region=us-central1 --quiet
GKE Skill Routing Table
Load the single, most specific GKE sub-skill below matching your workload
requirements. Do not load multiple GKE skills unless explicitly required.
| Scenario |
Trigger Keywords |
Target Skill |
| Golden Path Defaults |
production defaults, |
gke-golden-path |
| : : golden path : : |
|
|
| Cluster Creation |
create cluster, |
gke-cluster-creation |
| : : provision GKE : : |
|
|
| Networking & Ingress |
private cluster, VPC, |
gke-networking, |
: : Gateway API, Ingress, : gke-service-networking : |
|
|
| : : DNS : : |
|
|
| Security & IAM |
Workload Identity, |
gke-platform-security, |
: : Secret Manager, RBAC, : gke-workload-security : |
|
|
| : : hardening : : |
|
|
| Autoscaling |
HPA, VPA, Cluster |
gke-workload-scaling |
| : : Autoscaler, NAP : : |
|
|
| Compute Classes |
ComputeClass, Spot |
gke-compute-classes |
| : : fallback, GPU/TPU nodes : : |
|
|
| Cost Analysis |
BigQuery billing |
gke-cost-analysis |
| : : exports, budgets, live : : |
|
|
| : : monitoring : : |
|
|
| Cost Optimization |
Spot VMs, rightsizing, |
gke-cost-optimization |
| : : quotas : : |
|
|
| AI/ML Workloads |
LLM, GPU/TPU inference, |
gke-inference |
| : : serving, vLLM : : |
|
|
| GPU/TPU Disruption |
GPU termination, TPU |
gke-ai-troubleshooting- |
: : shutdown, host : handle-disruption-gpu-tpu : |
|
|
| : : maintenance : : |
|
|
| Cluster Upgrades |
upgrade, maintenance |
gke-upgrades |
| : : window, release channel : : |
|
|
| Observability |
monitoring, logging, |
gke-observability |
| : : Prometheus, dashboards : : |
|
|
| Multi-tenancy |
namespace isolation, |
gke-multitenancy |
| : : resource quota, : : |
|
|
| : : LimitRange : : |
|
|
| Batch & HPC |
batch, HPC, Kueue, |
gke-batch-hpc |
| : : JobSet, parallel jobs : : |
|
|
| App Onboarding |
containerize, |
gke-app-onboarding |
| : : Dockerfile, deploy app, : : |
|
|
| : : onboard : : |
|
|
| Backup & DR |
backup plan, restore, |
gke-backup-dr |
| : : disaster recovery, CMEK : : |
|
|
| Storage & PVC |
SSD, PV, PVC, |
gke-storage |
| : : StorageClass, GCS FUSE : : |
|
|
| Reliability |
PDB, health probe, |
gke-reliability |
| : : liveness, readiness : : |
|
|
| Productionization |
production readiness, |
gke-productionize |
| : : productionize, : : |
|
|
| : : readiness scoring, : : |
|
|
| : : audit cluster : : |
|
|
| Manifest Generation |
generate YAML, manifest |
gke-manifest-generation |
| : : template, : : |
|
|
| : : securityContext probes, : : |
|
|
| : : resource limits : : |
|
|
Conceptual & Informational Queries (CRITICAL)
For purely conceptual, educational, or informational questions (e.g. "What is
GKE?", "Explain GKE architecture", or "Compare Standard vs Autopilot" in a
generic sense):
- Rule: Answer immediately using your pre-trained knowledge.
- Constraint: Do not execute code searches, directory listings, or other
tool calls unless the user explicitly requests you to inspect the local
workspace or run a command. Keep it fast, cheap, and direct.
1---2name: gke-basics3description: Set up and manage Google Kubernetes Engine clusters, node pools, workloads, networking, and storage with Autopilot defaults.4---56**Trigger**: Use when working with Google Kubernetes Engine basics — cluster setup, node pools, workloads, networking, and storage fundamentals.78# GKE Basics910Managed Kubernetes platform on Google Cloud. Defaults to Autopilot mode.1112## Quick Start1314```bash15gcloud services enable container.googleapis.com --quiet16gcloud container clusters create-auto my-cluster --region=us-central1 --quiet17gcloud container clusters get-credentials my-cluster --region=us-central1 --quiet18```1920## GKE Skill Routing Table2122Load the single, most specific GKE sub-skill below matching your workload23requirements. **Do not load multiple GKE skills unless explicitly required.**2425| Scenario | Trigger Keywords | Target Skill |26| -------------------- | ----------------------- | --------------------------- |27| Golden Path Defaults | production defaults, | `gke-golden-path` |28: : golden path : :29| Cluster Creation | create cluster, | `gke-cluster-creation` |30: : provision GKE : :31| Networking & Ingress | private cluster, VPC, | `gke-networking`, |32: : Gateway API, Ingress, : `gke-service-networking` :33: : DNS : :34| Security & IAM | Workload Identity, | `gke-platform-security`, |35: : Secret Manager, RBAC, : `gke-workload-security` :36: : hardening : :37| Autoscaling | HPA, VPA, Cluster | `gke-workload-scaling` |38: : Autoscaler, NAP : :39| Compute Classes | ComputeClass, Spot | `gke-compute-classes` |40: : fallback, GPU/TPU nodes : :41| Cost Analysis | BigQuery billing | `gke-cost-analysis` |42: : exports, budgets, live : :43: : monitoring : :44| Cost Optimization | Spot VMs, rightsizing, | `gke-cost-optimization` |45: : quotas : :46| AI/ML Workloads | LLM, GPU/TPU inference, | `gke-inference` |47: : serving, vLLM : :48| GPU/TPU Disruption | GPU termination, TPU | `gke-ai-troubleshooting-` |49: : shutdown, host : `handle-disruption-gpu-tpu` :50: : maintenance : :51| Cluster Upgrades | upgrade, maintenance | `gke-upgrades` |52: : window, release channel : :53| Observability | monitoring, logging, | `gke-observability` |54: : Prometheus, dashboards : :55| Multi-tenancy | namespace isolation, | `gke-multitenancy` |56: : resource quota, : :57: : LimitRange : :58| Batch & HPC | batch, HPC, Kueue, | `gke-batch-hpc` |59: : JobSet, parallel jobs : :60| App Onboarding | containerize, | `gke-app-onboarding` |61: : Dockerfile, deploy app, : :62: : onboard : :63| Backup & DR | backup plan, restore, | `gke-backup-dr` |64: : disaster recovery, CMEK : :65| Storage & PVC | SSD, PV, PVC, | `gke-storage` |66: : StorageClass, GCS FUSE : :67| Reliability | PDB, health probe, | `gke-reliability` |68: : liveness, readiness : :69| Productionization | production readiness, | `gke-productionize` |70: : productionize, : :71: : readiness scoring, : :72: : audit cluster : :73| Manifest Generation | generate YAML, manifest | `gke-manifest-generation` |74: : template, : :75: : securityContext probes, : :76: : resource limits : :7778## Conceptual & Informational Queries (CRITICAL)7980For purely conceptual, educational, or informational questions (e.g. "What is81GKE?", "Explain GKE architecture", or "Compare Standard vs Autopilot" in a82generic sense):8384* **Rule**: **Answer immediately using your pre-trained knowledge.**85* **Constraint**: **Do not execute code searches, directory listings, or other86 tool calls** unless the user explicitly requests you to inspect the local87 workspace or run a command. Keep it fast, cheap, and direct.