The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external data files which presents a surface for indirect prompt injection.
Ingestion points: Data is loaded from local files using load_methylation_data, load_bed_file, and load_probe_annotation functions in SKILL.md.
Boundary markers: The skill does not implement specific boundary markers or instructions to ignore potential natural language commands embedded within the data files.
Capability inventory: The skill's functionality is limited to data processing and statistical analysis using pandas, numpy, and scipy. No hazardous capabilities such as arbitrary code execution or system-level modification were detected.
Sanitization: Input is parsed using structured data loaders like pd.read_csv, but the content is not sanitized for potential natural language instructions.- [EXTERNAL_DOWNLOADS]: The skill integrates with various scientific data services via the tooluniverse package.
The skill fetches genomic annotations and regulatory information from well-known scientific sources including Ensembl, ENCODE, SCREEN, and ChIPAtlas. These sources are recognized as safe and reputable in the bioinformatics domain, and the retrieval is consistent with the skill's intended use.

tooluniverse-epigenomics