data-labeling

The skill is benign and proportionate to its stated purpose. It describes legitimate data-labeling workflows, references standard tools, and does not exhibit credential harvesting, suspicious download behavior, or aggressive data exfiltration patterns. Some care should be taken when sharing code samples to avoid embedding placeholder credentials, and to ensure proper secure configuration of labeling tools in real deployments.