Holoscan Install Wheel
NVIDIAOfficialCreated by Holoscan TeamDistributed by NVIDIA
- Skill
- Workflow
- Software development
- Apache-2.0
- Medium risk
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
This skill provides a structured procedure for installing Holoscan SDK 5.x Python wheels, specifically designed for CUDA 13 environments. It emphasizes prerequisite validation, including system architecture, glibc versions, and driver compatibility, to ensure a stable installation. The workflow covers selecting the appropriate distribution packages, managing isolated virtual environments, and performing post-installation verification to confirm the SDK is correctly configured for development.
What it does
- Validates system prerequisites including CUDA 13 driver requirements and glibc versions.
- Guides the selection of Holoscan SDK 5.x distribution packages such as core, vision, or inference modules.
- Provides a step-by-step process for setting up an isolated Python virtual environment.
- Executes dry-run and installation commands for specific SDK versions.
- Performs post-installation verification checks including dependency validation and basic import tests.
- Outlines the process for running sample pipelines to confirm runtime functionality.
When to use it
- When setting up a new development environment for Holoscan SDK 5.x.
- When upgrading or installing specific Holoscan components in a clean Python environment.
- When troubleshooting installation issues related to CUDA 13 or library dependencies.
- When verifying that a Holoscan installation is correctly configured for Python-based applications.
Advantages
- Ensures environment isolation by mandating the use of virtual environments.
- Reduces installation errors through explicit prerequisite and compatibility checks.
- Provides a clear path for verifying both core and optional SDK modules.
- Standardizes the installation process for consistent development setups.
Usage examples
- “Install the full Holoscan 5.x SDK stack in a new virtual environment.”
- The agent will guide you through creating a venv, checking for CUDA 13 compatibility, and installing the holoscan package using pip.
- “How do I verify my Holoscan 5.x installation?”
- The agent will provide commands to run pip check, list installed packages, and execute a Python script to import the holoscan module and verify its version.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/nvidia/holoscan-install-wheel/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip holoscan-install-wheel.zip -d .agents/skills
Documentation by Holoscan Team
Holoscan 5.x Python Wheel Installation
Release and platform checks
This skill targets Holoscan SDK 5.x. Select the requested 5.x release before
choosing artifacts. Read its release notes and the matching
SDK installation guide.
Select the 5.x documentation version; an unversioned page may describe 4.x.
For an unreleased checkout, use its docs/sdk_installation.mdx and DEVELOP.md.
Early Access binary commands describe planned artifacts, so confirm publication
for the exact release before installing. If unavailable, explain that status
and offer holoscan-install-source; keep the requested 5.x version.
The 5.0 artifact set targets CUDA 13 and dGPU systems on x86_64 or arm64.
Native packages use Ubuntu 24.04; wheels use glibc 2.39 and regular CPython
3.12–3.14. Check the selected release for changes to that matrix. There is no
CUDA 12 variant or separate Tegra/iGPU artifact in this set. A driver reporting
CUDA 12 support needs a compatibility assessment or driver upgrade before a
CUDA 13 installation; the nvidia-smi CUDA field is a driver capability, not
the installed toolkit version. Never infer compatibility from that field alone.
Choose a distribution
Wheels contain Python bindings, typing information, and native runtime libraries. C++ headers, CMake files, examples, and sample data are separate. For native C++ development use Debian packages or a source build.
| Goal | Distribution |
|---|---|
| Full stack | holoscan (selects holoscan-cu13) |
| Core only | holoscan-core-cu13 |
| Optional modules | holoscan-vision-cu13, holoscan-visualization-cu13, holoscan-inference-cu13 |
Inspect the host, python3 --version, and getconf GNU_LIBC_VERSION. Confirm
regular CPython 3.12–3.14, glibc 2.39 or newer, a compatible Linux architecture,
and the CUDA 13 driver/runtime prerequisites. An active Conda environment does
not imply a supported Conda package. Prefer a separate venv for this workflow.
If these prerequisites are unmet, report the failed check and stop before
installation. Compare the driver version against the selected release's
documented CUDA 13 driver requirement; a toolkit version alone is insufficient.
Confirm the exact published 5.x distribution version on PyPI or the selected release's documented index; reserved package names are not installable SDKs. If absent, offer source instead of resolving an unpinned package to 4.x. Keep all components on the same SDK version.
Install and verify
If an installation or verification command fails, stop and report the failing command and output. Do not continue to the example or report success after a failed prerequisite, package-index, dry-run, install, dependency, or import check.
Choose a fresh venv path, or reuse an environment the user selected after checking its contents. Explain the chosen packages before installing:
python3 -m venv ~/holoscan/venv-5x
source ~/holoscan/venv-5x/bin/activate
# SDK_VERSION is the exact published Python distribution version for 5.x.
python -m pip index versions holoscan
python -m pip install --dry-run "holoscan==${SDK_VERSION}"
python -m pip install "holoscan==${SDK_VERSION}"
python -m pip check
python -m pip list --format=freeze
python -c 'import holoscan as hs; print(hs.__version__); print(hs.RealtimeClock); print(len(hs.registry.operators()))'For Core-only use holoscan-core-cu13 in the index, dry-run, and install
commands. Add optional component wheels at the exact matching SDK version.
Vision resolves CV-CUDA and NPP Python dependencies. The host supplies the
driver and CUDA 13 runtime; Inference expects TensorRT 10 in the application
environment. Verify optional providers separately when selected:
holoscan_vision, holoscan_visualization, and holoscan_inference.
For execution coverage, obtain examples/first_pipeline/python/first_pipeline.py
from the source ref corresponding to this release. Inspect it before execution
and run it in the venv; it is bounded and needs no RacerX data or display.
Prefer an existing matching checkout. Do not construct a Git tag from a Python
version string: Early Access packaging versions and source refs can differ.
The 4.x hello_world.py, tensor_interop, and video_replayer smoke sequence
does not describe the 5.x Python examples.
Report the Python, SDK, and component versions; import, dependency, and graph checks; and any skipped module tests. An import check alone is not full runtime qualification. Missing runtime libraries should be traced to the selected component's prerequisites before changing unrelated system paths.
Files
- ATTRIBUTION.mdAdded by AutoGineer931 B
- BENCHMARK.md8.2 KiB
- LICENSELicence file added by AutoGineer11.2 KiB
- SKILL.md5.0 KiB
- evals/evals.json6.5 KiB
- skill-card.md4.7 KiB
- skill.oms.sig4.5 KiB
evals/evals.json
[
{
"id": "holoscan-install-wheel-5x-001",
"question": "Review this request and return the installation decision and a command plan only. Host and publication facts below are test fixtures, not claims about this sandbox. Use the supplied facts and local skill guidance; do not execute commands, contact services, install packages, pull images, build, or run examples. Identify any missing prerequisite that must be checked before execution. Request: Install the published Holoscan 5.x full Python SDK in a venv on Ubuntu 24.04, Python 3.12, and a compatible CUDA 13 dGPU.",
"expected_skill": "holoscan-install-wheel",
"expected_script": null,
"ground_truth": "Produce a plan without executing it. Use an exact published 5.x metapackage version in an isolated environment.",
"expected_behavior": [
"Return the decision, prerequisite checks, and proposed commands without executing installations or external lookups; use the supplied host and release facts as fixtures.",
"Check Python, glibc, driver, CUDA runtime, and release availability.",
"Install the pinned holoscan metapackage and inspect component versions.",
"Run pip check and verify the 5.x Python API.",
"Use the matching first_pipeline source for bounded execution coverage."
]
},
{
"id": "holoscan-install-wheel-5x-002",
"question": "Review this request and return the installation decision and a command plan only. Host and publication facts below are test fixtures, not claims about this sandbox. Use the supplied facts and local skill guidance; do not execute commands, contact services, install packages, pull images, build, or run examples. Identify any missing prerequisite that must be checked before execution. Request: I need only Core and Inference for Holoscan 5.x on a supported arm64 CUDA 13 server.",
"expected_skill": "holoscan-install-wheel",
"expected_script": null,
"ground_truth": "Produce a plan without executing it. Install matching Core and Inference component wheels.",
"expected_behavior": [
"Return the decision, prerequisite checks, and proposed commands without executing installations or external lookups; use the supplied host and release facts as fixtures.",
"Use holoscan-core-cu13 and holoscan-inference-cu13 with the same version.",
"Check regular CPython 3.12\u20133.14 and glibc 2.39 requirements.",
"Check TensorRT 10 prerequisites and selected provider imports.",
"Do not claim wheels provide C++ headers or examples."
]
},
{
"id": "holoscan-install-wheel-5x-003",
"question": "Review this request and return the installation decision and a command plan only. Host and publication facts below are test fixtures, not claims about this sandbox. Use the supplied facts and local skill guidance; do not execute commands, contact services, install packages, pull images, build, or run examples. Identify any missing prerequisite that must be checked before execution. Request: My workstation has Python 3.10, glibc 2.35, and a CUDA 12-only driver. Install Holoscan 5.x via pip.",
"expected_skill": "holoscan-install-wheel",
"expected_script": null,
"ground_truth": "Produce a plan without executing it. Identify platform blockers before attempting package installation.",
"expected_behavior": [
"Return the decision, prerequisite checks, and proposed commands without executing installations or external lookups; use the supplied host and release facts as fixtures.",
"Explain the Python, glibc, and CUDA 13 requirements.",
"Do not choose holoscan-cu12 or fall back to 4.x.",
"Offer a compatible host or an assessed alternative instead of forcing installation."
]
},
{
"id": "holoscan-install-wheel-5x-004",
"question": "Review this request and return the installation decision and a command plan only. Host and publication facts below are test fixtures, not claims about this sandbox. Use the supplied facts and local skill guidance; do not execute commands, contact services, install packages, pull images, build, or run examples. Identify any missing prerequisite that must be checked before execution. Request: The holoscan name exists on PyPI but our 5.x release wheels are not published yet.",
"expected_skill": "holoscan-install-wheel",
"expected_script": null,
"ground_truth": "Produce a plan without executing it. A registered name is not proof of an installable 5.x release.",
"expected_behavior": [
"Return the decision, prerequisite checks, and proposed commands without executing installations or external lookups; use the supplied host and release facts as fixtures.",
"Check exact distribution versions.",
"Do not perform an unpinned install that could resolve to 4.x or a placeholder.",
"Offer a source build while retaining the 5.x target."
]
},
{
"id": "holoscan-install-wheel-5x-005",
"question": "Install native C++ Holoscan headers with apt.",
"expected_skill": null,
"expected_script": null,
"ground_truth": "Route to Debian rather than Python wheels.",
"expected_behavior": [
"Do not invoke the wheel installation skill for native C++ files."
]
},
{
"id": "holoscan-install-wheel-5x-006",
"question": "Review this request and return the installation decision and a command plan only. Host and publication facts below are test fixtures, not claims about this sandbox. Use the supplied facts and local skill guidance; do not execute commands, contact services, install packages, pull images, build, or run examples. Identify any missing prerequisite that must be checked before execution. Request: The requested Holoscan 5.0 pip install exited 1 with a dependency-resolution error. An older SDK is still importable in the venv and pip check returns 0. Can we continue to first_pipeline and call the 5.0 installation successful?",
"expected_skill": "holoscan-install-wheel",
"expected_script": null,
"ground_truth": "Stop after the failed requested-version installation and report its command and error. Successful checks against an older environment do not establish a successful 5.0 install.",
"expected_behavior": [
"Use only the supplied fixtures and local skill guidance; return a plan without executing commands or contacting services.",
"Report the failed installation command and supplied dependency-resolution error.",
"Do not proceed to first_pipeline or declare the requested 5.0 installation successful.",
"Do not silently substitute the older SDK or repair unrelated system paths."
]
}
]
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Needs Python
Security
What our scanner noticed· Medium risk
- Large base64-encoded blobs
- Instructions that use the network
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Created by
- Holoscan Team
- Publisher
- NVIDIA (Official)
- Distributed by
- NVIDIA
- Source
- NVIDIA/skills · skills/holoscan-install-wheel · commit fce3cca
- Licence
- Apache License 2.0 (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE-APACHE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Holoscan Team or NVIDIA.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download holoscan-install-wheel.zip