pdf-book-assembler

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/build_pdf_book.py uses subprocess.run to interact with system PDF utilities including pdfinfo, pdfseparate, pdfunite, gs, and pdftoppm. All commands are executed using list-based arguments without shell=True, which effectively prevents shell injection vulnerabilities.
  • [COMMAND_EXECUTION]: The script generates PostScript pdfmark instructions for Ghostscript to add bookmarks to the output PDF. It includes a dedicated escaping function escape_ps_string that sanitizes bookmark titles to prevent PostScript command injection.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external PDF files and a user-provided JSON specification. While this constitutes an attack surface for indirect prompt injection, the skill's logic is limited to page-level manipulation and metadata extraction. It does not process or interpolate PDF text content into agent prompts or execute it as code.
  • [SAFE]: The skill's Python script relies on standard third-party libraries numpy and Pillow for its whitespace detection logic, which are appropriate for the stated purpose. The script lacks network capabilities and does not attempt to access sensitive system files or establish persistence.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 07:34 AM