The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes several well-known third-party libraries for PDF manipulation across Python and JavaScript environments. These include Python packages pypdf, pdfplumber, reportlab, pytesseract, pdf2image, pandas, and pypdfium2, as well as JavaScript libraries pdf-lib and pdfjs-dist. These are standard tools for document processing.
[COMMAND_EXECUTION]: The skill integrates various command-line utilities from the Poppler and QPDF suites (such as pdftotext, pdftoppm, and qpdf) and provides custom Python scripts for specialized tasks. These scripts, such as scripts/fill_fillable_fields.py and scripts/extract_form_field_info.py, use standard subprocess patterns and include documented logic to handle PDF form structures, including a runtime monkeypatch of the pypdf library to resolve a known bug with selection lists.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted PDF documents.
Ingestion points: Untrusted data enters the agent's context through text extraction (using pypdf, pdfplumber, pdftotext) and image analysis of document pages (using pdf2image and visual inspection).
Boundary markers: No explicit delimiters or warnings are used when presenting extracted text or image analysis results to the agent.
Capability inventory: The skill has capabilities to read and write files on the local system and execute subprocesses via CLI tools and local Python scripts.
Sanitization: There is no evidence of sanitization or filtering of the content extracted from processed PDF documents before it is used by the agent.

pdf