The Agent Skills Directory

[SAFE]: The skill's primary function is to facilitate NGS analysis using the well-known deepTools suite. All documented workflows and parameters align with standard bioinformatics practices.
[SAFE]: The scripts/workflow_generator.py utility is implemented with a security-first approach to shell script generation:
It uses shlex.quote to properly escape all shell arguments, preventing command injection vulnerabilities.
It implements a SAFE_PATH_PATTERN regex and explicitly blocks directory traversal (e.g., '..') to ensure generated paths are restricted to the intended environment.
Numeric inputs are strictly validated using sanitize_positive_int.
[SAFE]: The scripts/validate_files.py utility performs thorough format validation for BAM, bigWig, and BED files. It includes defensive programming measures, such as safe header parsing for bigWig files to avoid potential library-level crashes.
[SAFE]: Dependencies are pinned to legitimate scientific packages (deepTools, pysam, pyBigWig) from official registries. References and citations point to established scientific sources (arXiv).
[INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection as it processes external genomics files (BAM, BED) that could theoretically contain malicious instructions in metadata or sequence fields. However, this risk is mitigated by the use of specialized binary parsers (pysam, pyBigWig) and a standalone validation script that inspects file structure before analysis begins. This capability is necessary for the skill's primary purpose and does not escalate the verdict.

deeptools