Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for using standard command-line tools such as pdftotext, qpdf, pdftk, and pdftoppm for document manipulation on local files.
- [EXTERNAL_DOWNLOADS]: The skill relies on established third-party libraries including pypdf, pdfplumber, reportlab, pytesseract, pdf2image, pandas, pypdfium2, pillow, and numpy.
- [DYNAMIC_EXECUTION]: The scripts/fill_fillable_fields.py script performs a targeted monkeypatch of the pypdf library at runtime to fix a specific bug in selection-list form field processing. This is a functional modification constrained to the local environment.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external PDF documents, creating a vector for indirect prompt injection if malicious content is extracted and subsequently processed by the agent.
- Ingestion points: PDF files are read and processed in extract_form_field_info.py, fill_fillable_fields.py, fill_pdf_form_with_annotations.py, and convert_pdf_to_images.py.
- Boundary markers: The workflow does not utilize explicit boundary markers or warnings to disregard instructions found within the document data.
- Capability inventory: The skill facilitates file system operations and shell command execution through the agent environment.
- Sanitization: No sanitization or content validation is performed on the data extracted from PDF documents.
Audit Metadata