The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the subprocess module in scripts/check_system.py to identify host hardware specifications.
Specifically, it executes sysctl -n hw.memsize and vm_stat on macOS to determine total and available RAM.
This functionality is benign and serves as a safeguard to prevent model loading on machines with insufficient memory.
[EXTERNAL_DOWNLOADS]: The skill is configured to download model weights from the official Google Research organization on Hugging Face Hub.
Resources include google/timesfm-2.5-200m-pytorch and google/timesfm-1.0-200m-pytorch.
These downloads are necessary for the core forecasting functionality and originate from a trusted organization.
[COMMAND_EXECUTION]: The script examples/global-temperature/run_example.sh is a shell utility for running demonstration code. It executes local Python scripts (run_forecast.py, visualize_forecast.py) and the system checker.

timesfm-forecasting