The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use various command-line utilities for PDF manipulation.
Evidence: The guide includes shell commands for pdftotext, qpdf, pdftk, and pdfimages to perform tasks like merging, splitting, and image extraction.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from external PDF files.
Ingestion points: Text and data are read from PDF files using pypdf.PdfReader, pdfplumber.open, and pytesseract in SKILL.md.
Boundary markers: Extracted text is handled directly without the use of delimiters or instructions to ignore instructions embedded within the PDF content.
Capability inventory: The skill can write files to the local system and execute external command-line tools.
Sanitization: The code snippets do not include logic to sanitize or validate the text extracted from the PDFs before it is used by the agent.

pdf