Pdf Cn

This skill implements PDF processing and form-filling code, but there are important inconsistencies you should address before trusting it with real data: - Dependency mismatch: The registry lists no required binaries or packages, but SKILL.md/scripts require system tools (pdftotext/pdftoppm/pdfimages/qpdf/pdftk, ImageMagick 'magick') and Python libraries (pypdf, pdfplumber, pdf2image, pytesseract, PIL, reportlab, etc.). Ensure you install these in a controlled environment (virtualenv or container) before running, and verify versions. - Licensing/provenance mismatch: LICENSE.txt claims restrictive Anthropic terms while _meta.json lists MIT and a different author/homepage. Confirm the license and source before redistributing or using in production. - Code audit: The included scripts perform local file I/O and modify PDFs (they do not perform network I/O or handle secrets), but you should still review the code and run it on non-sensitive example files first to validate behavior (especially the form-filling and coordinate transforms). - Run in isolation: Execute the scripts in a sandboxed environment (VM/container) and avoid processing sensitive documents until you are confident in the tool's outputs. If you need automatic agent-driven invocation, consider restricting it until dependencies and provenance are clarified. If these issues are resolved (declare and install dependencies, clarify license/source), the skill otherwise appears coherent with its stated purpose.