Manage GKE TPU Nodepool resources via XPK commands.
npx skills add https://github.com/primatrix/skills --skill apply-resource
Please help me install this Agent Skill. Skill: apply-resource Source: https://github.com/primatrix/skills/tree/main/plugins/exec-remote/skills/apply-resource Command: npx skills add https://github.com/primatrix/skills --skill apply-resource
This Skill automates the management of Google Kubernetes Engine (GKE) TPU Nodepool resources, streamlining the process of creating, deleting, listing, and describing TPU clusters.
To create a new GKE TPU cluster, use the command: /apply-resource create <cluster_name> <tpu_type> <num_slices> <zone>
High-intent search queries and answers about installing and using this skill.
It automates the management of GKE TPU Nodepool resources through XPK commands, handling the creation, deletion, listing, and describing of TPU clusters for machine learning workloads.
To provision GKE TPU clusters using XPK, you need gcloud, kubectl, and xpk installed. These dependencies interface with GKE to manage TPU resources directly from your environment.
To create a GKE TPU cluster with XPK, use the command `/apply-resource create <cluster_name> <tpu_type> <num_slices> <zone>`. This automates the TPU cluster provisioning process on GKE.
No, XPK does not cache GKE TPU cluster state locally. It queries GKE directly to ensure real-time cluster information and performs existence checks for multi-user environment safety.
Yes, you can use XPK to list and describe existing TPU clusters on GKE. It queries the most up-to-date cluster state directly from GKE to provide real-time cluster information.
Yes, XPK is safe for GKE TPU management in multi-user environments. It avoids local caching and performs existence checks to prevent conflicts when multiple users manage TPU resources.
primatrixPlan, render, and apply GKE TPU Job manifests on TPU v6e/v7x.
ssmleoAutomate batch and HPC workloads on GKE using Kubernetes jobs, JobSets, Kueue, and the MPI Operator.
cnoe-ioQuery EKS clusters via kubectl and report pod, node, and resource health.
fajshahProvision and manage multi-cloud GKE clusters with kubectl connectivity.
