Intel GPU AI Skills is a collection of agent skills for setting up, running, benchmarking, and profiling Hugging Face models on Intel GPUs. It supports workflows involving PyTorch, vLLM-XPU, SGLang-XPU, llama.cpp-SYCL, and migration from CUDA to XPU. The catalogue contains the project's skills, instructions, agent, and plugin.
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.
npx skills add intel/gpu-ai-skills --skill xpu-container-rungit clone --depth 1 https://github.com/intel/gpu-ai-skillsWrote 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.
[](https://agentmods.dev/skills/intel/gpu-ai-skills/xpu-container-run)<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/xpu-container-run"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/xpu-container-run/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/xpu-container-run"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/xpu-container-run.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 22 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 Privilege Escalation · line 3 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 19 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 20 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 44 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 56 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 106 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 17 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 54 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 70 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 90 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 42 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 61 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 71 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 93 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 98 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 74 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 79 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- high Privilege Escalation · line 123 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- medium MCP Rug Pull · line 3 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
- medium MCP Rug Pull · line 17 Docker image references without a specific tag (:latest is implicit) or digest (@sha256:...) can be silently replaced by a malicious image.Fix: Pin the image: image:tag or image@sha256:abc123
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00146 | $0.01690 |
| Opus 5 | $0.00073 | $0.00845 |
| Sonnet 5 | $0.00029 | $0.00338 |
| Haiku 4.5 | $0.00015 | $0.00169 |
Grade A, and why
xpu-container-run scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
xpu-container-run
Intel GPUs do not plug into Docker via --gpus all. There is no
nvidia-container-toolkit equivalent. Pass the kernel's Direct
Rendering Manager (DRM) character devices into the container and
grant the right group ownership.
CUDA → Intel cheat sheet
| CUDA | Intel |
|---|---|
docker run --gpus all |
--device /dev/dri --group-add "$(getent group render | cut -d: -f3)" |
docker run --gpus '"device=0"' |
-e ZE_AFFINITY_MASK=0 |
--ipc=host |
same |
--shm-size=16g |
same (alternative to --ipc=host) |
--runtime nvidia |
nothing — xe/i915 is in-kernel |
nvidia-smi inside container |
xpu-smi discovery |
No "Intel container toolkit" needed; passing the DRM nodes is enough.
Image source
Comes from the runner skill:
- vLLM serving → vllm-xpu-run (
vllm/vllm-openai-xpu:latest) - SGLang → sglang-xpu-run (built from upstream
docker/xpu.Dockerfile) - PyTorch / Transformers → torch-xpu-run
<image> below is whichever you picked.
Quickstart — interactive shell, one GPU
Confirm the image name with the user before running — this binds host GPU devices into the container.
docker run --rm -it \
--device /dev/dri \
--group-add "$(getent group render | cut -d: -f3)" \
--ipc=host \
-e ZE_AFFINITY_MASK=0 \
-e HF_TOKEN="$HF_TOKEN" \
-v "$HOME/.cache/huggingface:/root/.cache/huggingface" \
--entrypoint /bin/bash \
<image>
| Flag | Why |
|---|---|
--device /dev/dri |
Pass every Intel GPU's DRM nodes. Use --device /dev/dri/renderD128 for just the first GPU's render node (least privilege). |
--group-add "$(getent group render | cut -d: -f3)" |
Joins the container user to the host's render group by GID (not name) so it works in images where a render group with a different GID — or no render group at all — exists. Required when nodes are mode 0660/0640. Skip causes EACCES on Level Zero init. |
--ipc=host |
vLLM and torch.distributed use /dev/shm and POSIX semaphores. --shm-size=16g is a private-IPC alternative. |
-e ZE_AFFINITY_MASK=0 |
Pin to GPU 0. See xpu-discover for IDs. Always set explicitly. |
-v ~/.cache/huggingface:... |
Share the host model cache; avoid re-download. |
--entrypoint /bin/bash |
Override server-image autostart for interactive use. |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 127 lines · 146 tokens per session scan A 156b37202ef1
xpu-container-run is a skill published in the GitHub repository intel/gpu-ai-skills (21 stars, last pushed 4d ago), licensed Apache-2.0. It adds 146 tokens to every session and 1,690 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
holoscan-install-container
Install Holoscan SDK via the NGC Docker container. Use for container-based installs; not for native apt/pip/Conda installs.
doca-urom-svc
Operate the DOCA UROM Service container on BlueField Arm for remote memory operations (puts, gets, atomics, collectives) enqueued by a paired host using doca-urom: pull the NGC image, choose the UCX component, size queues, configure Comch pairing, and align host and service versions. SECURITY: the service has no…
hsb-setup
Clone the latest NVIDIA Holoscan Sensor Bridge repo, ask which supported devkit is being used, configure the host per platform, build the correct demo container, run it, and verify HSB connectivity by pinging 192.168.0.2. Use for Holoscan Sensor Bridge setup, build, container launch, and first-connectivity bring-up.
holoscan-install-source
Build Holoscan SDK from source via the in-tree ./run script. Use only when published packages don't meet the user's needs.
together-kueue
Install and use the Kueue job-queueing controller on a Together AI Kubernetes GPU cluster to gate jobs on quota. Covers installing Kueue, defining ResourceFlavor, ClusterQueue, and LocalQueue quota, submitting jobs to a queue, and watching quota admit or suspend them. Reach for it when a Together cluster's GPU pool…
edge-computing-expert
Design and implement production-grade edge computing solutions including CDN optimization, fog computing architectures, and distributed edge workloads. Use when the user mentions edge computing, CDN or edge caching, fog computing, MEC, IoT edge workloads, or latency-sensitive distributed deployments.