How do I install the Intel Gpu Device Selection skill?

Run npx skillmds@latest add tomevault-io/intel-gpu-device-selection in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Intel Gpu Device Selection safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Intel Gpu Device Selection?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Intel Gpu Device Selection free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Intel Gpu Device Selection?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Intel Gpu Device Selection by tomevault-io

Name: Intel Gpu Device Selection
Author: tomevault-io

Intel Gpu Device Selection

Select the Intel GPU device to use when a system has multiple Intel GPU devices. Use this skill when the user wants to run a workload on Intel GPU, mentions device selection, ZE_AFFINITY_MASK, or when multiple Level Zero GPU devices are detected and one must be chosen. Use when this capability is needed.

tomevault-io Updated Sep 17, 2026

File contents

Intel GPU Device Selection

Select a single Intel GPU device via ZE_AFFINITY_MASK when multiple Level Zero GPU devices are present.

This skill does NOT set ZE_FLAT_DEVICE_HIERARCHY. On PVC, the default is FLAT (each tile is a separate device). On other platforms, the default is COMPOSITE (tiles merged into one device). If the user needs to override this, they must set it explicitly before running this skill.

Instructions

Step 1: Detect Level Zero GPU devices

sycl-ls | grep -i level_zero

If sycl-ls is not available, use the "source-oneapi" skill first.

Only consider lines matching [level_zero:gpu][level_zero:<index>].

Step 2: Select the target device

0 devices found: Stop. Report: "No Level Zero GPU device was found. Check GPU driver, Level Zero runtime, and oneAPI environment."
1 device found: Use its index. Proceed to Step 3.
Multiple devices found: If the list contains both discrete and integrated GPUs, select the discrete GPU by default. If there are multiple discrete GPUs, ask the user to select. Do NOT proceed until a valid index is determined.

To distinguish dGPU from iGPU, run the bundled script:

python3 .claude/skills/action/intel-gpu-device-selection/scripts/l0_igpu_check.py

Output labels each GPU as iGPU or dGPU based on the Level Zero ZE_DEVICE_PROPERTY_FLAG_INTEGRATED flag.

Prompt format when user selection is needed:

Multiple Level Zero GPU devices found:
<device list from sycl-ls>

Which device should be used? Enter the device index (e.g., 0):

Step 3: Set ZE_AFFINITY_MASK

export ZE_AFFINITY_MASK=<selected index>

Source: intel/torch-xpu-ops — distributed by TomeVault.

---name: intel-gpu-device-selectiondescription: Select the Intel GPU device to use when a system has multiple Intel GPU devices. Use this skill when the user wants to run a workload on Intel GPU, mentions device selection, ZE_AFFINITY_MASK, or when multiple Level Zero GPU devices are detected and one must be chosen. Use when this capability is needed.---# Intel GPU Device SelectionSelect a single Intel GPU device via `ZE_AFFINITY_MASK` when multiple Level Zero GPU devices are present.This skill does NOT set `ZE_FLAT_DEVICE_HIERARCHY`. On PVC, the default is `FLAT` (each tile is a separate device). On other platforms, the default is `COMPOSITE` (tiles merged into one device). If the user needs to override this, they must set it explicitly before running this skill.## Instructions### Step 1: Detect Level Zero GPU devices```bashsycl-ls | grep -i level_zero```If `sycl-ls` is not available, use the "source-oneapi" skill first.Only consider lines matching `[level_zero:gpu][level_zero:<index>]`.### Step 2: Select the target device- **0 devices found**: Stop. Report: "No Level Zero GPU device was found. Check GPU driver, Level Zero runtime, and oneAPI environment."- **1 device found**: Use its index. Proceed to Step 3.- **Multiple devices found**: If the list contains both discrete and integrated GPUs, select the discrete GPU by default. If there are multiple discrete GPUs, ask the user to select. Do NOT proceed until a valid index is determined.To distinguish dGPU from iGPU, run the bundled script:```bashpython3 .claude/skills/action/intel-gpu-device-selection/scripts/l0_igpu_check.py```Output labels each GPU as `iGPU` or `dGPU` based on the Level Zero `ZE_DEVICE_PROPERTY_FLAG_INTEGRATED` flag.Prompt format when user selection is needed:```textMultiple Level Zero GPU devices found:<device list from sycl-ls>Which device should be used? Enter the device index (e.g., 0):```### Step 3: Set ZE_AFFINITY_MASK```bashexport ZE_AFFINITY_MASK=<selected index>```---> Source: [intel/torch-xpu-ops](https://github.com/intel/torch-xpu-ops) — distributed by [TomeVault](https://tomevault.io).

Frequently asked questions