The Agent Skills Directory

[SAFE]: The skill provides instructions and execution patterns for genomic data analysis using the OmicVerse library. All operations are confined to standard scientific data processing tasks (integration, preprocessing, and prediction).
[EXTERNAL_DOWNLOADS]: The documentation mentions external data sources like cellxgene for reference datasets, which is standard practice in bioinformatics for retrieving public datasets.
[COMMAND_EXECUTION]: The acceptance test in assets/acceptance.json uses a shell-executed Python script to verify the label transfer logic. This script operates on synthetic data in a controlled environment and uses standard scientific libraries (numpy, pandas, anndata).
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or suspicious downloads from untrusted sources were detected. The skill relies on local library installations and specified dependencies like scvi-tools.

omicverse-reference-label-transfer