Dev & Engineering ✓ NVIDIA · Official cuda-quantumquantum-computinggpu-simulationqpu-hardwarepythonc++

CUDA-Q Quantum Onboarding

Guides developers from CUDA-Q installation to quantum kernels, GPU simulation, and real QPU execution.

FollowSkills review · FSRS-2.0
Not recommended
54/ 100 5-point scale 2.7 / 5
Trust17 / 25 · 3.4/5

The skill declares only Read, Glob, and Grep, with no malware, credential theft, or destructive default evident. It gives useful credential-handling guidance, recommends emulate=True before QPU submission, and identifies cloud/QPU external effects. Deductions reflect missing per-step user confirmation, explicit permission boundaries, data-flow disclosure, and rollback guidance for installation, cloud access, and real-hardware submission.

Reliability8 / 20 · 2.0/5

Routing, prerequisites, target selection, platform branches, and troubleshooting are broadly self-consistent. However, the referenced local documentation is not included in the supplied material, dependency versions are not pinned, and there are no executable tests or comprehensive diagnosable failure paths for the key workflows. Static evidence is therefore insufficient for a higher score.

Adaptability10 / 15 · 3.3/5

Audience and scenarios are clear across installation, test programs, GPU simulation, QPU access, applications, and parallelization, with several non-fit conditions documented. The evaluation guidance also distinguishes CUDA-Q from unrelated or near-match requests. Chinese-language interaction is not specified, and cloud/QPU guidance partly depends on external services without mainland-China reachability alternatives, so points are deducted.

Convention9 / 15 · 3.0/5

The skill has structured sections, routing, examples, limitations, troubleshooting, version 1.0.1, author attribution, and an Apache-2.0 declaration; the skill card adds ownership information. It lacks a clear changelog, explicit maintenance/update path, and dependency version matrix. The supplied repository license metadata is NOASSERTION while local files state licenses, leaving governance-level licensing uncertainty.

Effectiveness6 / 15 · 2.0/5

The material can directly guide installation choices, Bell-state validation, simulator selection, QPU dialogue, application exploration, and multi-GPU parallelization. Nevertheless, the full installation and Bell program steps are not included, the workflow depends on absent documentation files, and static review cannot confirm that commands, APIs, or expected output work. The reported evaluation results are not independently reproducible from the supplied files, so the score remains near the static ceiling.

Verifiability4 / 10 · 2.0/5

A fixed revision, document paths, external documentation references, structured eval files, and a claimed passing benchmark provide limited traceability. There is no independently reproducible key-path test suite, CI coverage, or primary third-party result in the supplied material; the benchmark is principally an author-provided report, so the static score remains limited.

Evidence confidence:Low Reviewed Jul 20, 2026 Reviewed revision 55f18499943e
Before you use it
  • Do not treat the external documentation links, installation commands, or provider procedures as verified; confirm versions, targets, cost, permissions, and credential scope before installation, network access, or real-hardware submission.
  • For mainland-China users, verify reachability of NVIDIA documentation, cloud services, and QPU provider endpoints in advance, and prepare offline documentation or local-simulation alternatives.
  • Add a complete Bell-state example, a dependency/version matrix, command-level confirmation points, rollback guidance, and executable CI coverage to reduce documentation-drift risk.
See the full review method →

What does this skill do, and when should you use it?

This is NVIDIA’s onboarding skill for CUDA-Q, aimed at Python and C++ users. It covers installation, a Bell-state validation program, GPU-accelerated simulation, real QPU access, application examples, and multi-GPU or multi-QPU execution. Linux supports NVIDIA GPU targets, macOS is CPU-simulation oriented, and Windows requires WSL. Questions about compiler internals are routed to a separate compiler-focused skill.

Reads local CUDA-Q documentation and routes requests to installation, test programs, GPU simulation, QPU access, applications, or parallelization; provides pip and C++ installation commands; builds and runs a Bell-state kernel and checks for output near { 00:~500 11:~500 }; explains cudaq.sample(), run(), observe(), and get_state(); recommends nvidia, tensornet, qpp-cpu, mgpu, or mqpu targets; and walks users through QPU setup by technology and provider.

  1. A Python beginner installing CUDA-Q locally and running a first Bell-state program.
  2. A developer with an NVIDIA GPU choosing among single-GPU, multi-GPU, or tensor-network simulation backends.
  3. A macOS or GPU-less user configuring CPU-only simulation.
  4. A quantum application developer exploring QAOA, VQE, Grover’s algorithm, quantum machine learning, or finance examples.
  5. A user preparing access to providers such as IonQ, Quantinuum, IQM, QuEra, or AWS Braket.
  6. A developer running many circuits concurrently or distributing a large Hamiltonian calculation across GPUs.

What are this skill's strengths and limitations?

Pros
  • Covers the CUDA-Q path from installation through QPU access.
  • Supports both Python and C++ and distinguishes CPU, GPU, multi-GPU, and multi-QPU workflows.
  • Includes a concrete validation step, credential-handling guidance, and troubleshooting advice.
  • The skill is Apache-2.0 licensed; the catalog repository is maintained by NVIDIA and synced daily from product repositories.
Limitations
  • GPU simulation requires Linux; macOS is CPU-only and Windows requires WSL.
  • Kernel Python is restricted, and NumPy or SciPy cannot be used inside kernels.
  • The `mgpu` target requires MPI, while QPU access requires provider accounts and credentials.
  • The source does not provide an independent test suite or practical platform-test evidence for this individual skill.

How do you install this skill?

Install the skill from NVIDIA’s collection with the default CLI: npx skills add nvidia/skills --skill cudaq-guide --yes. The README does not specify the exact local installation path; cloning the repository or copying folders manually is not required. Update installed skills with npx skills update.

How do you use this skill?

Invoke it with /cudaq-guide; without an argument it displays the full menu. Direct routes include /cudaq-guide install, /cudaq-guide test-program, /cudaq-guide gpu-sim, /cudaq-guide qpu, /cudaq-guide applications, and /cudaq-guide parallelize. The QPU flow first asks for a technology category and then a provider.

How does this skill compare with similar options?

Compared with qpp-cpu, the nvidia target provides NVIDIA GPU-accelerated simulation, while qpp-cpu is the CPU-only fallback for GPU-less systems, macOS, or small test circuits. mgpu pools memory for one large circuit; mqpu runs independent circuits in parallel.

FAQ

Does this skill cost money to use?
The source specifies Apache-2.0 licensing for the skill; it does not state whether QPU services incur fees.
Can it run on macOS or Windows?
macOS Apple Silicon supports CUDA-Q with CPU simulation; Windows requires WSL followed by the Linux procedure.
Why might a CUDA-Q kernel fail to compile?
Check that the `@cudaq.kernel` body uses supported Python constructs; NumPy and SciPy are not allowed inside kernels.
How should QPU credentials be protected?
Use environment variables or an uncommitted local profile, and do not place credentials in source code, notebooks, logs, or commits.

More skills from this repository

All from NVIDIA/skills

Automation & Ops ✓ NVIDIA · Official

cuOpt Installation Guide

Install and verify NVIDIA cuOpt for Python, C, or REST deployment.

Dev & Engineering ✓ NVIDIA · Official

CuTile Autotuning Guide

Design, implement, and validate low-overhead autotuning for CuTile GPU kernels.

Dev & Engineering ✓ NVIDIA · Official

Holoscan Python Wheel Installer

Install Holoscan SDK Python bindings in a virtual environment and verify them on a Linux NVIDIA GPU system.

Dev & Engineering ✓ NVIDIA · Official

DALI Dynamic Mode Assistant

Helps agents write, review, and migrate NVIDIA DALI imperative dynamic-mode code.

Dev & Engineering ✓ NVIDIA · Official

NeMo Relay Migration Assistant

Safely migrate NeMo Flow projects to NeMo Relay.

Dev & Engineering ✓ NVIDIA · Official

Holoscan SDK Setup Guide

Inspects a Linux host and selects the most suitable Holoscan SDK installation path.

Dev & Engineering ✓ NVIDIA · Official

Earth2Studio Deterministic Forecast

Build single-member weather forecast inference scripts with Earth2Studio.

Dev & Engineering ✓ NVIDIA · Official

Catheter Navigation Fluoroscopy Smoke Tests

Validate core catheter-navigation workflow behavior on CPU-only CI environments.

Dev & Engineering ✓ NVIDIA · Official

NeMo Relay Installation Guide

Choose and verify the right NeMo Relay path for CLIs, language packages, and maintained frameworks.

Dev & Engineering ✓ NVIDIA · Official

NeMo Relay Quick Start

Helps first-time NeMo Relay users prove observable execution value through the smallest suitable trial.

Data & Analysis ✓ NVIDIA · Official

DICOM Series Preflight

Header-only validation for one DICOM series before conversion or inference.

Dev & Engineering ✓ NVIDIA · Official

Earth2Studio Diagnostic Builder

Build Earth2Studio wrappers for single-step diagnostic data transformations.

Data & Analysis ✓ NVIDIA · Official

Earth2Studio Weather Data Fetch

Fetch validated weather and climate variables from Earth2Studio sources by time and data type.

Data & Analysis ✓ NVIDIA · Official

NVIDIA AI-Q Deep Research

Run deep research through a reachable local or self-hosted AI-Q Blueprint backend.

Dev & Engineering ✓ NVIDIA · Official

AMC Sample Dataset Calibration

Verify a running NVIDIA AutoMagicCalib service end to end with its bundled sample dataset.

Dev & Engineering ✓ NVIDIA · Official

NeMo Relay Typed Wrappers & Codecs

Add typed boundaries to NeMo Relay integrations while preserving predictable JSON middleware semantics and caller-visible behavior.

Dev & Engineering ✓ NVIDIA · Official

Omniverse Realtime Viewer

Routes USD viewer requests to the right architecture and guides rendering, interaction, UI, and validation.

Dev & Engineering ✓ NVIDIA · Official

Clinical ASR Flywheel: Environment Setup

Validate that a clinical ASR evaluation environment can complete a TTS-to-ASR round trip through NVIDIA-hosted speech services.

Data & Analysis ✓ NVIDIA · Official

TAO DAFT Dataset Converter

Guides AI agents through tao-daft conversion between supported NVIDIA TAO DAFT dataset formats.

Dev & Engineering ✓ NVIDIA · Official

AutoMagicCalib Video Calibration

Turn local multi-camera MP4 recordings into AutoMagicCalib calibration results through a REST API.

Related skills