The Agent Skills Directory

[PROMPT_INJECTION]: The skill facilitates the extraction of text and metadata from user-provided PDF files, creating a potential vector for indirect prompt injection where malicious content within a PDF could influence the agent's behavior.
Ingestion points: Text extraction via pypdf and pdfplumber in SKILL.md, and OCR extraction via pytesseract in REFERENCE.md.
Boundary markers: The provided scripts and code snippets do not implement specific boundary markers or 'ignore' instructions to isolate extracted document content from system prompts.
Capability inventory: The skill includes the ability to write files to the disk and execute various command-line PDF processing utilities.
Sanitization: There is no evidence of sanitization or content filtering on the text data extracted from external PDF files.
[COMMAND_EXECUTION]: The documentation and reference guides provide instructions and examples for executing standard command-line tools for PDF operations, including qpdf, pdftotext, and pdfimages.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and standard third-party libraries for document processing (e.g., reportlab, pypdfium2, pdf-lib) that are downloaded from official package registries.

pdf