Does it support Apple Silicon GPUs?

Yes, it specifically detects M1, M2, M3, and M4 chips, identifies Metal support, and checks for unified memory availability.

What libraries does it suggest for parallel processing?

Depending on your core count, it provides strategies for using joblib, multiprocessing, Dask, and Zarr to maximize throughput.

Is a specific Python package required?

Yes, the detection script requires the psutil library to accurately read system memory and process information.

What system resources does this skill detect?

It identifies CPU core counts and architecture, GPU availability (NVIDIA, AMD, Apple Silicon), total and available RAM, and remaining disk space.

Why should I run this at the start of a project?

Running it early prevents Out-of-Memory (OOM) errors and ensures you choose the right libraries (like Dask for large data) before committing to a specific architecture.

Scientific Compute Resource Detector

Name: Scientific Compute Resource Detector
Author: x-cmd

byx-cmd

•

データサイエンスとML

Analyzes system hardware to generate optimized strategic recommendations for computationally intensive scientific and machine learning tasks.

This skill proactively assesses a system's CPU, GPU, memory, and disk availability to inform architecture decisions before starting resource-heavy operations. By generating a detailed JSON report and tailored recommendations, it helps AI agents and developers choose the most efficient processing strategy—such as selecting between parallel processing libraries like Dask and joblib, or determining when to leverage GPU acceleration via CUDA, ROCm, or Apple Silicon Metal. It ensures that large-scale data analysis, model training, and simulations are executed within the constraints of the local environment to prevent crashes and optimize performance.

主な機能

01Real-time monitoring of available RAM and disk space constraints

02Comprehensive GPU analysis for NVIDIA (CUDA), AMD (ROCm), and Apple Silicon (Metal)

03Context-aware recommendations for parallel processing and memory management

048 GitHub stars

05Standardized JSON export for automated decision-making in downstream scripts

06Automated detection of CPU cores, architecture, and frequency specifications

ユースケース

01Determining whether a massive dataset requires out-of-core processing or can fit in RAM

02Calculating the optimal number of parallel workers for multi-core processing tasks

03Detecting available hardware backends to automatically configure PyTorch or JAX models

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add x-cmd/skill get-available-resources

For use in Claude.ai and ChatGPT

Download Skill