The Agent Skills Directory

[SAFE]: The skill instructions and associated Python scripts do not contain any external network requests or remote resource downloads. All operations are designed to be performed on local data files provided via command-line arguments.
[DATA_EXFILTRATION]: The skill includes scripts/check_access_level.py, a utility specifically designed to audit data flow. It categorizes data by sensitivity (raw, redacted, verified) and prevents files marked with high sensitivity from being directed toward public-facing 'sinks' such as research papers or public repositories.
[COMMAND_EXECUTION]: Python scripts utilize standard library modules like argparse to handle local file system paths. The processing logic is focused on statistical analysis and data transformation using established libraries such as pandas and numpy. There is no evidence of arbitrary command injection or execution of untrusted system commands.
[SAFE]: The scripts/quality_gate.py tool uses yaml.safe_load to process configuration files, which is a security best practice that prevents unsafe deserialization and potential remote code execution via malicious YAML files.
[SAFE]: The skill follows 'least privilege' principles by providing localized tools that perform specific data engineering tasks without requiring elevated system permissions or broad network access.

light-data-engineering