hyperpod-version-checker

Name: hyperpod-version-checker
Author: awslabs

github.com/awslabsCloud & DevOps932146 installsUpdated 3w ago

About hyperpod-version-checker

hyperpod-version-checker is a script-based skill for checking software and hardware-related components on HyperPod nodes. It produces text or JSON reports that can be compared across nodes, and checks CUDA and driver compatibility.

Key features

01Check GPU and CUDA components
02Inspect EFA, NCCL, MPI, and GDRCopy
03Check Neuron SDK and ML software
04Inspect container runtimes
05Output text or JSON reports
06Analyze CUDA and driver compatibility

Use cases

01Audit component versions on a node
02Compare software stacks across cluster nodes
03Check driver and CUDA compatibility

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/awslabs/agent-plugins --skill hyperpod-version-checker

After installing, type/hyperpod-version-checkerin Claude Code to confirm it loaded.

Bundled filesscripts/hyperpod_check_versions.sh

Details

From the project README

HyperPod Version Checker

Upload to cluster nodes via hyperpod-ssm skill, then execute.

Usage

# Text report to console + file
bash hyperpod_check_versions.sh

# JSON only to stdout (text report still saved to file) — best for piping/parsing
bash hyperpod_check_versions.sh --json

# Custom output file
bash hyperpod_check_versions.sh --output /tmp/versions.txt

# No color (for logging)
bash hyperpod_check_versions.sh --no-color

Output file: component_versions_<hostname>_<timestamp>.txt (default)

What It Checks

Component	Detection Method	Applicable When
NVIDIA Driver	`nvidia-smi`	GPU instances (p3/p4/p5/g5)
CUDA Toolkit	`nvcc`, `/usr/local/cuda` symlink	GPU instances
cuDNN	Header file, packages	GPU instances doing deep learning
NCCL	Library filename, header, packages	Distributed GPU training
EFA	`/opt/amazon/efa_installed_packages`, `fi_info`	EFA-capable instances (p4d/p4de/p5/trn1/trn2)
AWS OFI NCCL	`efa_installed_packages`, library search	EFA + NCCL workloads
GDRCopy	rpm/dpkg, kernel module	GPU instances with RDMA (p4d+/p5)
MPI	`mpirun`, `/opt/amazon/openmpi`	Distributed training
Neuron SDK	`neuronx-cc`, `neuron-ls`, packages	Trainium/Inferentia (trn1/trn2/inf1/inf2)
Python/PyTorch	`python3`, `torch` import	ML workloads
Container runtime	`docker`, `containerd`, `kubectl`, `nvidia-ctk`	EKS clusters

Multi-Node Comparison

Run on each node individually via the hyperpod-ssm skill. With --json, stdout is clean JSON for easy diffing.

Compatibility Reference

The script automatically analyzes CUDA/driver compatibility. For reference:

Driver Series	Supported CUDA
580+	13.x, 12.x, 11.x
570+	12.8+ (Blackwell), 12.x, 11.x
545+	12.3-12.7, 11.x
525-535	12.0-12.2, 11.x
450+	11.x only

NCCL: Use 2.18+ for CUDA 12.x, 2.12+ for CUDA 11.x. Must be consistent across all nodes.

EFA Installer	AWS OFI NCCL
1.29+	v1.7.3+ (recommended)
1.26-1.28	v1.7.0-v1.7.2
1.20-1.25	v1.6.0+

Comments

Markdown supported

Details

Author: awslabs
Repository: awslabs/agent-plugins
Category: Cloud & DevOps
Installs: 146
GitHub stars: 932

Report

More skills

All skills

lark-apps

open.feishu.cn

732.0k

Develop and host Miaoda (Spark/Miaoda) apps: app creation, local full-stack development, cloud generation and iteration, creative design (UI mockups / interactive prototypes / wireframes / landing pages / dashboards / slide decks / visual exploration), integration of AI-related capabilities, Feishu platform capabilities, or other external capabilities; log/Trace/monitoring metric/PV/UV queries; environment variable management; app collaborators and collaboration permissions; app roles and members; and automation triggers (scheduled / record changes / Webhook / Feishu approvals). Use when users want to develop or create a system, tool, platform, or app; develop locally or in the cloud, modify, deploy, publish, launch, or get a shareable link; create a page or website with HTML and deploy it to Miaoda; design, create a mockup, prototype, wireframe, PPT, deck, or visual exploration; or mention Miaoda/Spark/Miaoda apps (links like `/app/app_xxx` or `/page/<token>`, runtime domains like *.aiforce.cloud), app databases, app file storage, open API Keys, visibility, app collaborators/developer permissions, app roles/role members, production logs, API request volume, error volume, latency, traffic, environment variables, or setting up automation tasks, scheduled triggers, or approval-triggered automation for Miaoda apps. Not for regular cloud drive file uploads (lark-drive), Feishu document editing (lark-doc), or native slide creation (lark-slides).

azure-prepare

github.com/microsoft

571.7k

Prepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wants to use azd as the deployment tool, or the project already has an azure.yaml file. DO NOT USE FOR: non-azd deployments, Python App Service code-only deploys (use python-appservice-deploy), or cross-cloud migration (use azure-cloud-migrate). WHEN: prepare app for azd, create azure.yaml, set up azd infrastructure, modernize app for Azure with azd, deploy with azd, function app, timer trigger, service bus trigger, event-driven function, managed identity, generate Bicep, generate Terraform, create and deploy to Azure.

wizard

github.com/mattpocock

571.5k

Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.

azure-ai

github.com/microsoft

571.2k

Use for Azure AI: Search, Speech, OpenAI, Document Intelligence. Helps with search, vector/hybrid search, speech-to-text, text-to-speech, transcription, OCR. WHEN: AI Search, query search, vector search, hybrid search, semantic search, speech-to-text, text-to-speech, transcribe, OCR, convert text to speech.