The Agent Skills Directory

[SAFE]: The skill relies on well-known and reputable Python libraries such as pypdf, pdfplumber, reportlab, and Pillow for its core functionality. It includes a specific monkeypatch in scripts/fill_fillable_fields.py to address a known bug in the pypdf library related to selection list processing, which is a benign and documented developer workaround.
[COMMAND_EXECUTION]: The documentation in SKILL.md and reference.md provides examples for executing standard PDF command-line utilities (e.g., qpdf, pdftotext, pdfimages, pdftk). These tools are part of common Linux/macOS environments and are used here for legitimate document manipulation tasks.
[PROMPT_INJECTION]: The skill processes untrusted external data (PDF files), which introduces a risk of indirect prompt injection.
Ingestion points: External text and metadata enter the agent's context through pypdf (in scripts/extract_form_field_info.py), pdfplumber, and OCR via pytesseract (documented in SKILL.md).
Boundary markers: No explicit delimiters or boundary markers are used in the scripts' output to separate extracted content from agent instructions.
Capability inventory: The skill includes file system read/write access for documents and images, and references multiple CLI tools for document processing.
Sanitization: The skill does not perform sanitization, filtering, or validation on the text extracted from the PDF documents before it is presented to the agent.

pdf