The Agent Skills Directory

[SAFE]: The skill performs expected bioinformatics workflows using local scripts and standard libraries. No unauthorized network operations, credential harvesting, or safety bypass attempts were detected.
[SAFE]: The use of Base64 encoding in scripts/step7_report.py is for legitimate image embedding within the generated HTML analysis reports.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface, loading files (CSV, H5AD, FCS) in scripts/step1_load_data.py. It has capabilities including run_terminal and write_file. While it lacks explicit sanitization and boundary markers for the data, the risk is considered safe as the ingested content is strictly numerical biological data used for statistical processing and does not influence the agent's control flow or instructions.

bioinformatics-init-analysis