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-profile-unitracegit 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-profile-unitrace)<a href="https://agentmods.dev/skills/intel/gpu-ai-skills/xpu-profile-unitrace"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/xpu-profile-unitrace/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-profile-unitrace"><img src="https://agentmods.dev/badge/skills/intel/gpu-ai-skills/xpu-profile-unitrace.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 137 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.
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 | $0.00127 | $0.01612 |
| Opus 5 | $0.00063 | $0.00806 |
| Sonnet 5 | $0.00025 | $0.00322 |
| Haiku 4.5 | $0.00013 | $0.00161 |
Grade A, and why
xpu-profile-unitrace 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 10d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
xpu-profile-unitrace
unitrace profiles XPU workloads at the SYCL / Level Zero kernel
level — captures per-kernel timing, memcopy bytes, oneCCL/MPI
events, and hardware counters that PyTorch profilers can't see.
Use when:
- A PyTorch-level profiler identified a hot op and you need to know which SYCL kernel inside it is the cost.
- You need Level Zero command-list events, oneCCL collectives, exact memcopy bytes, kernel launch geometry, or HW counters.
- Multi-XPU run with per-rank oneCCL visibility needed.
- Workload is raw SYCL / oneAPI (not PyTorch).
Prefer torch-xpu-profile or vllm-xpu-profile first; their output usually answers the question without going to SYCL level.
Install: check, then build if missing
command -v unitrace && unitrace --version
Most public XPU images don't ship unitrace. To build from source inside the target image:
source /opt/intel/oneapi/setvars.sh --force >/dev/null
git clone --depth 1 https://github.com/intel/pti-gpu.git /opt/pti-gpu
cd /opt/pti-gpu/tools/unitrace
mkdir build && cd build
cmake -DCMAKE_BUILD_TYPE=Release ..
make -j"$(nproc)"
export PATH="/opt/pti-gpu/tools/unitrace/build:$PATH"
unitrace --version # confirm
unitrace --device-list # confirm sees XPU
Prerequisites: CMake 3.22+, C++17 compiler, oneAPI Base Toolkit
(present in vllm/vllm-openai-xpu:latest and any sglang-xpu image).
Add -DBUILD_WITH_MPI=1 for multi-GPU collective profiling;
-DCMAKE_INSTALL_PREFIX=/opt/unitrace && make install for an
installable layout. Verified clean on vllm/vllm-openai-xpu:latest
against unitrace 2.3.0.
If you profile often, bake this into a Dockerfile extending the runtime image so you don't rebuild every session.
Quickstart capture
cd /work # working dir is where the trace lands
unitrace \
--chrome-call-logging \
--chrome-kernel-logging \
python3 my_workload.py
Closing log lines name the exact paths:
[INFO] Log is stored in /work/python3.<PID>.json
[INFO] Timeline is stored in python3.<PID>.json
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.
- 10d ago First seen · 159 lines · 127 tokens per session scan A 8d1386de5797
xpu-profile-unitrace is a skill published in the GitHub repository intel/gpu-ai-skills (21 stars, last pushed 5d ago), licensed Apache-2.0. It adds 127 tokens to every session and 1,612 once invoked, about $0.0006 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
gke-compute-classes
Configures, optimizes, and troubleshoots GKE ComputeClasses. Use when configuring Spot VMs with on-demand fallback, targeting specific accelerators (GPUs/TPUs) or machine families, restricting ComputeClass access, or debugging pending pods related to node pool auto-creation. Do not use for cluster-level Node Auto…
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.
doca-socket-relay
Use this skill when the operator is driving the DOCA Socket Relay to bridge a socket-oriented host application onto a BlueField DPU peer without rewriting it — picking the deployment shape (in-process, sidecar, or BlueField service container), configuring the host-side socket and the DPU-side forwarding endpoint…
offensive-z-wave
Z-Wave attack methodology — sniffing with Z-Force / EZ-Wave / RTL-SDR + ZniffMobile, S0 (legacy) network-key derivation flaw and key reuse, S2 (modern) ECDH commissioning analysis, replay/injection on unauthenticated nodes, default-key brute-force on test deployments, and home-automation hub pivots. Use when targeting…
hsb-flash
Flash the FPGA on an HSB board connected to an NVIDIA devkit. Supports HSB Lattice boards (FPGA versions 2407, 2412, 2507, 2510) and Leopard Imaging VB1940 "all-in-one" cameras (FPGA versions 2507, 2510). Uses release-specific YAML manifests and board-type-specific program commands. Lattice and VB1940 commands must…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.