The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it instructs the agent to extract and analyze content from user-provided PDF files.
Ingestion points: The agent uses tools like pypdf, pdfplumber, and pytesseract (OCR) to extract text and image data from external PDF files (as seen in scripts/extract_form_field_info.py and scripts/convert_pdf_to_images.py).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between the skill's own instructions and instructions potentially embedded within the data of the processed PDFs.
Capability inventory: The skill allows for the execution of shell commands (qpdf, pdftotext) and local file system writes.
Sanitization: No sanitization or filtering is performed on the extracted text or OCR results before they are presented to the agent for analysis in forms.md and SKILL.md.
[COMMAND_EXECUTION]: The skill documentation and scripts facilitate the execution of several command-line utilities for PDF processing.
Evidence: SKILL.md and reference.md provide command-line examples for pdftotext, qpdf, pdftk, pdfimages, and pdftoppm to perform document transformations and data extraction.

pdf