Does this skill support Apple Silicon hardware?

Yes, it specifically detects M1/M2/M3/M4 chips and recommends Metal-specific backends (MPS) for libraries like PyTorch, TensorFlow, and JAX.

How does it help with large datasets?

It compares available RAM to your dataset size and recommends out-of-core strategies like Dask, Zarr, or H5py if memory is constrained.

Where is the resource information stored?

The skill generates a '.claude_resources.json' file in your current working directory, which can be read programmatically by your other scripts.

Are there any external dependencies required?

The detection script requires the 'psutil' Python library, which can be installed via 'uv pip install psutil' or standard pip.

Scientific Compute Resource Analyzer

Name: Scientific Compute Resource Analyzer
Author: pur3v4d3r

bypur3v4d3r

•

Data Science & ML

Detects and analyzes system hardware to provide optimized strategic recommendations for scientific computing and data processing tasks.

This skill proactively identifies available CPU cores, GPU hardware (NVIDIA, AMD, Apple Silicon), memory, and disk space to inform high-level architectural decisions for computationally intensive work. By generating a structured JSON report with strategic recommendations, it helps developers and data scientists choose the most efficient approach—such as parallel processing with Dask, out-of-core computing, or specific GPU acceleration backends—before initiating resource-heavy model training or data analysis.

Key Features

01Automated detection of CPU, GPU (CUDA, ROCm, Metal), RAM, and Disk availability

02Generates structured JSON output for programmatic integration in scripts

03Library-specific suggestions for PyTorch, JAX, Dask, and joblib

04Context-aware strategic recommendations for parallel and distributed computing

05Cross-platform support for macOS (Apple Silicon), Linux, and Windows

061 GitHub stars

Use Cases

01Determining the ideal number of parallel workers for high-concurrency batch processing

02Configuring optimal GPU backends for machine learning model training and inference

03Optimizing data loading strategies for large-scale datasets that exceed system RAM

Key Features

01Automated detection of CPU, GPU (CUDA, ROCm, Metal), RAM, and Disk availability

02Generates structured JSON output for programmatic integration in scripts

03Library-specific suggestions for PyTorch, JAX, Dask, and joblib

04Context-aware strategic recommendations for parallel and distributed computing

05Cross-platform support for macOS (Apple Silicon), Linux, and Windows

061 GitHub stars

Use Cases

01Determining the ideal number of parallel workers for high-concurrency batch processing

02Configuring optimal GPU backends for machine learning model training and inference

03Optimizing data loading strategies for large-scale datasets that exceed system RAM