The Agent Skills Directory

[SAFE]: The skill is focused on scientific analysis and metadata extraction. The code in lib/extractor.py performs legitimate PDF parsing using well-known libraries such as PyMuPDF (fitz) and pdfplumber.
[EXTERNAL_DOWNLOADS]: The skill specifies dependencies on standard, reputable Python packages (pymupdf, pdfplumber, numpy) which are necessary for its documented functionality.
[COMMAND_EXECUTION]: No dangerous command execution or dynamic code evaluation patterns were found. The extraction logic is purely text and image processing based on regular expressions and PDF object inspection.
[DATA_EXFILTRATION]: There are no indicators of unauthorized data access or exfiltration. The skill operates on user-provided PDFs or fetched academic papers from standard sources like arXiv.

sci-extract