Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implementation is consistent with its stated purpose of PDF processing. No malicious code, persistence mechanisms, or unauthorized data access patterns were detected.
- [EXTERNAL_DOWNLOADS]: The skill references and utilizes multiple standard industry libraries for PDF and image handling, including
pypdf,pdfplumber,reportlab,pytesseract,pdf2image,pypdfium2,Pillow,pandas,numpy,pdf-lib, andpdfjs-dist. These are appropriate for the skill's functionality. - [COMMAND_EXECUTION]: The documentation provides instructions for using established command-line tools such as
qpdf,pdftotext,pdftk, andmagick(ImageMagick). These are intended for local execution by the agent to perform specific PDF transformations. - [PROMPT_INJECTION]: The skill processes untrusted external PDF files, creating a surface for indirect prompt injection where malicious instructions could be embedded in document content or metadata.
- Ingestion points: Text and layout extraction via
scripts/extract_form_structure.pyandscripts/extract_form_field_info.py. - Boundary markers: Absent; the skill does not explicitly wrap extracted content in delimiters.
- Capability inventory: The skill includes capabilities for file system writes and subprocess execution via shell commands.
- Sanitization: The scripts rely on library-level parsing (e.g.,
pypdf,pdfplumber) and do not implement additional sanitization of extracted text before it is returned to the agent context.
Audit Metadata