Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard documentation and helper scripts for processing PDF documents using well-known libraries like
pypdf,pdfplumber,reportlab,pdf2image, andpdf-lib. - [COMMAND_EXECUTION]: The documentation includes standard usage examples for common command-line utilities such as
qpdf,pdftotext, andpdftk. These are utility-specific commands within the expected scope of the skill's purpose. - [DYNAMIC_EXECUTION]: The script
scripts/fill_fillable_fields.pycontains a monkeypatch for thepypdflibrary'sDictionaryObject.get_inheritedmethod to fix a specific bug in version 5.7.0. This is a local, controlled modification of a library function to ensure correct functionality and does not introduce external or untrusted code execution. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external PDF files, which is an inherent ingestion point for untrusted data. However, the instructions emphasize manual and automated validation steps (e.g.,
check_bounding_boxes.py, visual inspection of validation images) to ensure that data extraction and form filling are accurate and under user/agent control. Standard library guardrails for PDF parsing are in place.
Audit Metadata