get-available-resources

To check available system resources for Python computations, you can run a detection script that reads CPU cores, memory, disk space, and GPU data. The script outputs a .claude_resources.json file containing specific recommendations for parallelism, memory strategy, and GPU backend usage.

You should detect system resources for computational planning at project initiation or immediately before executing intensive tasks. This resource detection determines your available constraints, allowing you to proactively decide on appropriate parallel processing, memory management, and GPU backend usage.

Yes, the resource detection script supports NVIDIA CUDA, ROCm, and Apple Metal GPUs. It reads GPU data alongside CPU, memory, and disk constraints, generating a .claude_resources.json file that provides guidance on which GPU backend to use for your computational tasks.

To generate a JSON file with hardware recommendations for parallel processing, run the resource detection script at the start of your workflow. It analyzes your CPU, memory, disk, and GPU data to produce a .claude_resources.json file containing actionable usage guidance.

Yes, you need the psutil library installed as a dependency to detect system memory and disk constraints. The Python script uses psutil to read available CPU, memory, and disk resources, which are then combined with GPU data to output computational planning recommendations.

The best way to decide between CPU and GPU backends for heavy computation is to run a system resource detection script. By reading your available CPU, memory, disk, and GPU constraints, it generates a .claude_resources.json file with tailored recommendations for backend usage and parallelism.