The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided PDF documents to identify text anchors for content placement, which is a common attack surface for document-processing tools. \n
Ingestion points: The skill reads PDF content using pymupdf.open() and page.get_textpage_ocr() as described in SKILL.md. \n
Boundary markers: No specific delimiters or safety instructions are provided to the agent to ignore potentially malicious embedded text. \n
Capability inventory: The skill utilizes file writing and document modification via the pymupdf library. \n
Sanitization: No sanitization of text labels found within the PDF is performed before using their coordinates. \n- [EXTERNAL_DOWNLOADS]: The skill requires standard installations of the pymupdf and pillow Python packages and the tesseract OCR engine, all of which are established tools for these tasks. \n- [COMMAND_EXECUTION]: The skill provides commands for environment setup, including pip install for Python dependencies and brew install for system tools like Tesseract.

pdf-sign