EvilFreelancer

GPU Server Setup — DevOps skill for Claude Code

DevOps community

Agent skill that prepares a Linux server with NVIDIA GPUs for LLM workloads - driver and CUDA, Docker with NVIDIA Container Toolkit, staged verification gates, compose presets for vLLM, Infinity, Open.

How to install GPU Server Setup

This entry records only its repository, not the path inside it, so there is no exact command to give. Open EvilFreelancer/gpu-server-setup and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What GPU Server Setup does

Alternatives in DevOps

Ship Release — Systematic release workflow for RTK: build verification, version bump, changelog update, git tag, and push to 11.9k ★
Skypilot — Run, manage, and scale AI workloads on any AI infrastructure 9.7k ★
Infra Deploy — Use when changing Terraform, Docker, AWS or Azure deployment assets, or deployment documentation, including te 47 ★

README

gpu-server-setup

An agent skill that prepares a **Linux server with NVIDIA GPUs** for neural-network and LLM workloads — from a bare Debian/Ubuntu install to a verified Docker + GPU inference stack. It is a portable skill: install it in Claude Code, Cursor, or OpenAI Codex, or drop the folder into any agent that reads `SKILL.md`.

Based on two field-tested sources by the author: the Dzen article ["Как подготовить Linux к запуску и обучению нейросетей? (+ Docker)"](https://dzen.ru/a/ZVt9kRBCTCGlQqyP) and a production setup runbook for multi-GPU inference servers.

What it does

NVIDIA driver + CUDA — from the official CUDA apt repo (cuda-keyring), with the open-vs-proprietary kernel-module choice, driver-only vs toolkit decision, multi-CUDA via update-alternatives, and version pinning.
Docker + NVIDIA Container Toolkit — official Docker Engine install, nvidia-ctk runtime configure, and explicit GPU passthrough (--gpus, compose deploy blocks, per-service device_ids pinning).
Staged verification — hard gates between layers (lspci → nvidia-smi → GPU-in-Docker → service health); the skill never proceeds past a failing gate. scripts/check.sh runs the gates automatically.
Inference presets (optional) — the mandatory deliverable is the verified Docker + GPU base; serving stacks go on top only when asked for. Ready docker-compose files for vLLM (per-GPU model services), Infinity (embeddings), OpenWebUI (web UI) and Ollama (llama.cpp runs the same way via its official image), with the flags that matter explained.
GPU power limit — on request: query the card's allowed range, cap with nvidia-smi -pl, verify the limit took effect, persist it across reboots with a systemd unit.
Troubleshooting — symptom → cause → fix table for the classic failures: Secure Boot vs DKMS, "Driver/library version mismatch", could not select device driver "nvidia", CUDA OOM, silent CPU fallback.