A specialized toolkit for managing and querying NVIDIA CUDA environments and GPU device information on Jetson platforms.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install jetson-cuda
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install jetson-cuda using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Jetson CUDA?
The Jetson CUDA skill is a dedicated utility designed for developers working on the NVIDIA Jetson ecosystem, including AGX Orin, Xavier, and Nano series. It streamlines the process of environment verification by providing instant access to CUDA toolkit versions and hardware specifications.
By integrating this tool into your Openclaw Skills library, you can ensure that your AI agent has the necessary context to troubleshoot driver issues, verify compiler versions via nvcc, and monitor GPU health through standard NVIDIA interfaces.
Jetson CUDA Use Cases
- Verifying CUDA compiler (nvcc) versions during AI model compilation.
- Checking GPU device availability and memory status on Jetson hardware.
- Locating CUDA library paths for linking in C++ or Python projects.
- Auditing environment compatibility for JetPack SDK deployments.
How Jetson CUDA Works
- The skill interfaces with the local Jetson Linux environment to locate the CUDA installation directory.
- It executes system-level commands like nvcc and nvidia-smi to extract versioning and hardware telemetry.
- The collected data is parsed and formatted into a structured output for the AI agent or developer.
- It identifies specific path configurations to ensure dependencies are correctly mapped for Openclaw Skills workflows.
Jetson CUDA Setup
Ensure your NVIDIA Jetson device is running a supported version of JetPack with the CUDA Toolkit pre-installed. You can initialize the skill using the following commands:
# Verify your current CUDA environment
jetson-cuda info
# Ensure nvcc is accessible in your PATH
which nvcc
nvcc --version
Jetson CUDA Data Schema & Taxonomy
The skill retrieves and organizes system metadata as follows:
| Data Point | Source Command | Description |
|---|---|---|
| CUDA Version | nvcc --version |
The current version of the CUDA compiler toolkit. |
| GPU Metrics | nvidia-smi |
Real-time hardware status, utilization, and driver version. |
| Library Paths | System Environment | The export paths for CUDA headers and shared objects. |
Jetson CUDA Advanced Features
- Support for multi-version CUDA environment detection (11.x and 12.x).
- Automated path validation for common Jetson directory structures.
- Seamless integration with other Openclaw Skills for end-to-end AI deployment on the edge.
SKILL.md
Loading
Related Openclaw Skills
Suno Poetry Music Creator
A sophisticated AI music engineering skill that transforms themes into polished Suno songs through style DNA analysis and iterative lyric refinement.
Website Structure Analyzer
A comprehensive auditing tool that deconstructs any website into its core functional modules, information architecture, and underlying technology stack.
Tianshu Homework Split
A specialized utility that transforms dense academic instructions and syllabus outlines into structured Markdown task lists.
HPC FEniCS
A specialized AI agent skill for building, debugging, and optimizing finite-element workflows using the FEniCS and DOLFINx stacks.
Deep Research
An advanced research pipeline that decomposes complex queries into multi-round searches to generate structured reports with verified citations.
QLD Property Research
An automated research tool for conducting comprehensive Queensland property due diligence using verified official government and council data sources.
Featured*
