一个代理技能，用于准备搭载NVIDIA或Intel Arc GPU的Linux服务器 用于神经网络和大型语言模型工作负载——从裸Debian/Ubuntu安装到一个 验证了Docker + GPU推理栈。这是一个便携技能：安装在 Claude Code、Cursor 或 OpenAI Codex，或者把文件夹放进任何代理 读数为SKILL.md。

基于作者经过实地验证的两个来源：Dzen文章 “Как подготовить Linux к запуску и обучению нейросетей？（+ Docker）” 以及多GPU推理服务器的生产环境运行手册。英特尔弧 部分原因是2026年在Debian 13上搭建了双Arc A770服务器。

它的作用

• NVIDIA 驱动 + CUDA — 来自官方 CUDA apt 仓库（cuda-keyring）， 在开放与专有内核模块选择中，仅驱动 VS 工具包 判定、通过update-alternatives进行多CUDA处理，以及版本钉定。
• 强制内核头 — 精确运行内核头加上跟踪 metapackage（linux-headers-amd64）在标准Debian上;匹配内核味道 其他地方），并验证了已安装内核的头部和NVIDIA模块 在重启之前。内核升级不能在下一次启动时无声禁用GPU。

gpu-server-setup

An agent skill that prepares a Linux server with NVIDIA or Intel Arc 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)” and a production setup runbook for multi-GPU inference servers. The Intel Arc part comes from bringing up a dual Arc A770 server on Debian 13 in 2026.

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.
• Mandatory kernel headers — exact running-kernel headers plus the tracking metapackage (linux-headers-amd64 on standard Debian; match the kernel flavour elsewhere), with headers and NVIDIA modules verified for installed kernels before a reboot. A kernel upgrade must not silently disable GPUs on the next boot.
• Intel Arc (Alchemist and Battlemage) — the in-tree i915/xe driver with distro firmware, Resizable BAR and real PCIe link checks, the trixie-backports kernel when the host must see GPU temperatures, Docker access through /dev/dri render nodes with no toolkit, and a field-tested map of what serves on Arc today: llama.cpp SYCL/Vulkan, Ollama, OpenVINO Model Server, vLLM on Battlemage and a no-build vLLM recipe for the A770, plus measured i915 against xe on a pair of A770: what the force-probed xe driver gives (up to 5x for vLLM, 2x for llama.cpp) and what it costs (a card that can wedge until a reboot, runtime PM to switch off).
• 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 or i915/xe + render nodes + GuC → GPU-in-Docker → service health); the skill never proceeds past a failing gate. scripts/check.sh detects the vendor and 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 or Intel hwmon power1_max, verify the limit took effect, persist it across reboots with a systemd unit (scripts/arc-power-cap.sh for Intel).