Apply Resource

• Install auth plugin: gke-gcloud-auth-plugin
• Check: kubectl version --client

Defaults

The following defaults apply unless the user explicitly overrides them:

Use these values directly — do NOT ask the user to confirm or re-enter them unless they specify otherwise.

Required Parameters

When creating a cluster, the following parameters are required:

• PROJECT_ID: GCP project ID (default: tpu-service-473302)
• CLUSTER_NAME: Name of the GKE cluster (default: sglang-jax-agent-tests)
• TPU_TYPE: TPU accelerator type (e.g., v6e-16, v6e-4, v4-8) — must be specified
• NUM_SLICES: Number of TPU slices to provision (default: 1)
• ZONE: GCP zone (default: asia-northeast1-b)

If these parameters are already known from an upstream caller (e.g., exec-remote or deploy-cluster), use them directly — do NOT re-ask the user. Only prompt interactively when this skill is invoked standalone and the user wants to override defaults.

Operations

1. Create Cluster

Creates a new GKE cluster with TPU nodepool using Pathways.

Command:

xpk cluster create-pathways \
  --cluster $CLUSTER_NAME \
  --num-slices=$NUM_SLICES \
  --tpu-type=$TPU_TYPE \
  --zone=$ZONE \
  --spot \
  --project=$PROJECT_ID

Interactive Flow:

1. Prompt user for CLUSTER_NAME
2. Prompt user for TPU_TYPE
3. Prompt user for NUM_SLICES
4. Prompt user for ZONE
5. Validate TPU availability in the specified zone
6. Check if cluster name already exists in GKE (prevents conflicts)
7. Execute cluster creation

2. Delete Cluster

Deletes an existing GKE cluster.

Command:

xpk cluster delete \
  --cluster $CLUSTER_NAME \
  --zone=$ZONE \
  --project=$PROJECT_ID

Interactive Flow:

1. Query GKE for all live clusters
2. Display cluster list with current status
3. Prompt user to select a cluster
4. Verify cluster still exists (multi-user safety check)
5. Confirm deletion
6. Execute cluster deletion

3. List Clusters

Lists all managed clusters.

Command:

xpk cluster list

Real-time Query:

• Queries GKE directly using gcloud container clusters list
• Shows current status, location, and node pool information
• No local cache - always up-to-date

4. Describe Cluster

Shows detailed information about a specific cluster.

Command:

xpk cluster describe \
  --cluster $CLUSTER_NAME \
  --zone=$ZONE

TPU Availability Handling

When a specified ZONE doesn't support the requested TPU_TYPE:

1. Automatically fetch available zones from TPU regions documentation
2. Suggest alternative zones that support the TPU type
3. Allow user to select a different zone or cancel

Multi-User Safety Features

This skill is designed for multi-user environments:

1. No Local Caching: All cluster information is queried from GKE in real-time
2. Existence Checks: Before creating, checks if cluster name already exists
3. Pre-deletion Verification: Before deleting, verifies cluster still exists
4. Location Validation: Automatically uses actual cluster location if different from user input
5. Concurrent Operation Safe: Multiple users can operate simultaneously without conflicts

Parameter Passing

This skill can be invoked in two ways:

1. Standalone (/apply-resource create): Prompt user for all required parameters interactively.
2. From pipeline (called by deploy-cluster or exec-remote): Parameters (CLUSTER_NAME, TPU_TYPE, NUM_SLICES, ZONE) are already known — use them directly without prompting.

Error Handling

• Missing prerequisites: Check if xpk, gcloud, kubectl are installed
• Authentication errors: Verify gcloud auth list
• Zone unavailability: Automatically suggest alternative zones
• Quota errors: Display quota information and suggest solutions
• Network errors: Retry with exponential backoff

Useful Resources

• TPU Architecture
• TPU Regions and Zones
• XPK Cluster Documentation