The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocesses to execute well-known command-line utilities for PDF processing, such as qpdf, pdftotext, pdftk, and ImageMagick (magick/convert), to perform operations like merging, rotating, and cropping PDF pages.
[EXTERNAL_DOWNLOADS]: Skill instructions and reference files suggest the installation of established Python and JavaScript libraries from official registries (PyPI, npm), including pypdf, pdfplumber, reportlab, pytesseract, pdf2image, pypdfium2, pdf-lib, and pdfjs-dist.
[PROMPT_INJECTION]: The skill processes untrusted PDF files, creating a surface for indirect prompt injection where malicious content inside a PDF could attempt to influence the agent's behavior during text analysis. * Ingestion points: PDF content is ingested via scripts/extract_form_structure.py, scripts/extract_form_field_info.py, and various text extraction patterns in SKILL.md. * Boundary markers: None identified. Extracted text is not delimited or separated from the main instructional context with warnings to ignore embedded commands. * Capability inventory: The skill has the ability to execute shell commands, write files to the system, and generate images from document pages. * Sanitization: No sanitization or filtering logic is present to clean extracted text content before it is processed by the agent.

pdf