The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to run local Python scripts including comprehensive_analysis.py, combine_sources.py, and genotype_cram.py, and to perform environment management tasks such as creating virtual environments.
[EXTERNAL_DOWNLOADS]: The skill performs environment setup by installing Python packages (including pandas, numpy, scipy, and reportlab) from the official PyPI registry using pip install -r requirements.txt.
[PROMPT_INJECTION]: The skill processes untrusted external data from genomic files and script-generated output (agent_summary.json), creating a surface for indirect prompt injection.
Ingestion points: User-provided DNA files (e.g., 23andMe, AncestryDNA, VCF) and analysis summaries.
Boundary markers: Absent in the instructions for processing external content.
Capability inventory: Bash for command execution, Read for file access, and SendUserFile for report generation.
Sanitization: No explicit sanitization or validation of external input data is provided in the instruction set.

personal-genomics