gpu-kubernetes-operations

gpu-kubernetes-operations is a skill for Claude Code, Codex from BagelHole/DevOps-Security-Agent-Skills. It costs 33 tokens per session (3,117 once invoked), scanned A, original, MIT.

A guide to running Kubernetes clusters with GPUs for AI model serving and training. It covers scheduling workloads, sharing GPUs, monitoring hardware, scaling nodes, and handling failures.

In plain words
What is it for?
Use it to set up NVIDIA GPU support, configure GPU node pools and MIG partitions, monitor them with Prometheus and DCGM, troubleshoot scheduling or out-of-memory errors, and control costs.
Why use it?
GPU clusters are harder to operate than ordinary servers because drivers, device allocation, memory limits, and hardware health all affect whether jobs run. This guide organizes those operational tasks and troubleshooting checks.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to set up NVIDIA GPU support, configure GPU node pools and MIG partitions, monitor them with Prometheus and DCGM, troubleshoot scheduling or out-of-memory errors, and control costs.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/bagelhole/devops-security-agent-skills/gpu-kubernetes-operations
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add BagelHole/DevOps-Security-Agent-Skills --skill gpu-kubernetes-operations
Clone the repo
git clone --depth 1 https://github.com/BagelHole/DevOps-Security-Agent-Skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for gpu-kubernetes-operations

README.md
[![agentmods](https://agentmods.dev/badge/skills/bagelhole/devops-security-agent-skills/gpu-kubernetes-operations.svg)](https://agentmods.dev/skills/bagelhole/devops-security-agent-skills/gpu-kubernetes-operations)
Your own site
<a href="https://agentmods.dev/skills/bagelhole/devops-security-agent-skills/gpu-kubernetes-operations"><img src="https://agentmods.dev/badge/skills/bagelhole/devops-security-agent-skills/gpu-kubernetes-operations.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,117 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Tool Misuse · line 86
    Code deploys a privileged Kubernetes workload (privileged container, hostPath mount, or host namespaces). This grants root on the node and is a node/cluster takeover vector.
    Fix: Remove privileged, hostPath, and host-namespace settings from workloads. Use a least-privilege securityContext, drop capabilities, and avoid mounting the host filesystem.
  • high Tool Misuse · line 99
    Code deploys a privileged Kubernetes workload (privileged container, hostPath mount, or host namespaces). This grants root on the node and is a node/cluster takeover vector.
    Fix: Remove privileged, hostPath, and host-namespace settings from workloads. Use a least-privilege securityContext, drop capabilities, and avoid mounting the host filesystem.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00033 $0.03117
Opus 5 $0.00016 $0.01558
Sonnet 5 $0.00007 $0.00623
Haiku 4.5 $0.00003 $0.00312

Measured 4d ago against content hash 18fc7be68d53, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

gpu-kubernetes-operations scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

infrastructure/local-ai/gpu-kubernetes-operations/SKILL.md · 453 lines

How it starts

The opening of the file, as written. The whole thing — 453 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GPU Kubernetes Operations

Run resilient and cost-efficient GPU clusters for production AI workloads.

When to Use This Skill

  • Setting up GPU node pools in Kubernetes for AI inference or training
  • Configuring NVIDIA device plugin and GPU operator
  • Implementing MIG partitioning to share GPUs across workloads
  • Building GPU-aware autoscaling policies
  • Monitoring GPU health with DCGM and Prometheus
  • Troubleshooting GPU scheduling, driver, or OOM issues

Prerequisites

  • Kubernetes 1.28+ cluster with GPU-capable nodes
  • NVIDIA GPUs (A10, L4, A100, H100, or similar)
  • NVIDIA drivers installed on nodes (535+ recommended)
  • Helm 3 for operator and plugin installation
  • Prometheus stack for metrics collection

NVIDIA GPU Operator Installation

The GPU Operator automates driver, toolkit, device plugin, and DCGM deployment.

# Add NVIDIA Helm repo
helm repo add nvidia https://helm.ngc.nvidia.com/nvidia
helm repo update

# Install GPU Operator
helm install gpu-operator nvidia/gpu-operator \
  --namespace gpu-operator \
  --create-namespace \
  --set driver.enabled=true \
  --set toolkit.enabled=true \
  --set devicePlugin.enabled=true \
  --set dcgmExporter.enabled=true \
  --set migManager.enabled=true \
  --set nodeStatusExporter.enabled=true \
  --version v24.3.0

# Verify installation
kubectl get pods -n gpu-operator
kubectl get nodes -o json | jq '.items[].status.allocatable["nvidia.com/gpu"]'

NVIDIA Device Plugin (Standalone)

If not using the GPU Operator, deploy the device plugin directly.

# nvidia-device-plugin.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: nvidia-device-plugin
  namespace: kube-system
spec:
  selector:
    matchLabels:
      name: nvidia-device-plugin
  template:
    metadata:
      labels:
        name: nvidia-device-plugin
    spec:
      tolerations:
        - key: nvidia.com/gpu
          operator: Exists
          effect: NoSchedule
      priorityClassName: system-node-critical
      containers:
        - name: nvidia-device-plugin
          image: nvcr.io/nvidia/k8s-device-plugin:v0.15.0
          securityContext:
            privileged: true
          env:
            - name: FAIL_ON_INIT_ERROR
              value: "false"
            - name: DEVICE_SPLIT_COUNT
              value: "1"
            - name: DEVICE_LIST_STRATEGY
              value: "envvar"
          volumeMounts:
            - name: device-plugin
              mountPath: /var/lib/kubelet/device-plugins
      volumes:
        - name: device-plugin
          hostPath:
            path: /var/lib/kubelet/device-plugins

Read the full file on GitHub · 453 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 453 lines · 33 tokens per session scan A 18fc7be68d53

Subscribe to this mod's changes

gpu-kubernetes-operations is a skill published in the GitHub repository BagelHole/DevOps-Security-Agent-Skills (1,058 stars, last pushed 3mo ago), licensed MIT. It adds 33 tokens to every session and 3,117 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

agentic-eks-bootstrap

Bootstrap an AWS EKS cluster optimized for Agentic AI workloads — Karpenter v1.2+ GPU node pools, EKS Auto Mode, Kubernetes 1.32+ with DRA 1.35 GA, VPC CNI, GPU Operator, and baseline observability. Use when starting a new EKS cluster that will host vLLM, Inference Gateway, Langfuse, or Kagent.

aws-samples/sample-oh-my-aidlcops · 91 tokens

gpu-resource-management

Design GPU orchestration on EKS using Karpenter v1.2+ NodePools, KEDA scale-to-zero, and DRA 1.35 GA for multi-instance GPU (MIG) partitioning. Right-size NodePool for p5/g6e/trn2 instance mix, spot/on-demand split, consolidation, and topology-aware scheduling.

aws-samples/sample-oh-my-aidlcops · 76 tokens

inference-gateway-routing

Configure kgateway v2.0+ as L1 and Bifrost v1.x or LiteLLM v1.60+ as L2 for a 2-Tier Inference Gateway on EKS. Apply Cascade Routing (Haiku→Sonnet→Opus fallback), Semantic Router (intent-based model pick), and HTTPRoute with OTel trace propagation to Langfuse.

aws-samples/sample-oh-my-aidlcops · 84 tokens

vllm-serving-setup

Design, deploy, and tune vLLM v0.18.2 inference serving on EKS with PagedAttention v2, Multi-LoRA, FP8 KV Cache, Chunked Prefill, and Continuous Batching. Produces Helm values.yaml, PodMonitor, HPA, and kubectl validation steps for production agentic workloads.

aws-samples/sample-oh-my-aidlcops · 76 tokens

aws-bedrock-ai

WORKFLOW SKILL — Amazon Bedrock and AWS AI design: foundation model selection, knowledge bases (RAG), agents for bedrock, guardrails, provisioned throughput, batch inference, fine-tuning, KMS, VPC endpoints, regional GA, and per-provider licensing.

odere-pro/claude-aws-architect · 65 tokens

implementing-aws-config-rules-for-compliance

Implementing AWS Config rules for continuous compliance monitoring of AWS resources, deploying managed and custom rules aligned to CIS and PCI DSS frameworks, configuring automatic remediation with SSM Automation, and aggregating compliance data across accounts.

adriannoes/awesome-agentic-ai · 53 tokens