The Agent Skills Directory

[SAFE]: The skill provides legitimate bioinformatics functionality for gene regulatory network inference. No malicious patterns, obfuscation, or unauthorized data access operations were detected during the analysis.
[COMMAND_EXECUTION]: The skill includes a standalone script scripts/basic_grn_inference.py for CLI execution. This is a standard and expected feature for processing large genomic datasets.
[EXTERNAL_DOWNLOADS]: References installation of 'arboreto' and its dependencies (dask, numpy, pandas, scikit-learn, scipy) from official registries like PyPI and Bioconda. These are well-known, trusted sources for scientific software.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external genomic data for analysis, creating a potential attack surface.
- Ingestion points: Expression data (TSV) and transcription factor files are loaded via pandas.read_csv and arboreto.utils.load_tf_names in scripts/basic_grn_inference.py.
- Boundary markers: Absent; the input is treated as structured tabular or list data for numerical computation.
- Capability inventory: File writing (to_csv) and network communication via Dask (for distributed computing) are used for legitimate data processing and scaling.
- Sanitization: Input is parsed into structured numeric formats (Pandas DataFrames, NumPy arrays) by the library, which inherently restricts the data from being interpreted as agent instructions.

arboreto