The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-supplied PDF documents through OCR and automated text extraction. This content is subsequently interpreted by the agent to identify form fields and enter data, creating an indirect prompt injection surface.
Ingestion points: External PDF files are processed in scripts/convert_pdf_to_images.py, scripts/extract_form_field_info.py, and scripts/check_fillable_fields.py.
Boundary markers: None identified; the agent is instructed to directly analyze the extracted text and images.
Capability inventory: The skill includes Python scripts for PDF creation and manipulation, and instructions for running system CLI tools.
Sanitization: No specific sanitization or filtering of extracted content is performed before presentation to the agent.
[COMMAND_EXECUTION]: The skill provides detailed instructions for the agent to execute various command-line tools such as qpdf, pdftotext, pdftk, and pdfimages. It also includes several Python utility scripts designed to be executed via the command line for tasks like field extraction and image conversion.
[EXTERNAL_DOWNLOADS]: The documentation references well-known and trusted external libraries including pypdf, pdfplumber, reportlab, and pdf-lib. These are industry-standard tools for PDF manipulation and are treated as safe dependencies.
[SAFE]: A monkeypatch is implemented in scripts/fill_fillable_fields.py to address a specific bug in the pypdf library (v5.7.0). This runtime modification is hardcoded and does not incorporate untrusted external input.

pdf