Pdf Anthropic

Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multipl...

This skill appears coherent and focused on PDFs, but note a few practical points before installing/using it: (1) There is no automated installer — you must have the Python packages and native tools the scripts use (pypdf, pdfplumber, pdf2image, pillow, poppler utilities, qpdf/pdftk, ImageMagick) installed on the host. (2) The code monkeypatches a pypdf DictionaryObject method to handle certain option formats — this is a compatibility change that could break with different pypdf versions; test on non-sensitive PDFs first. (3) The LICENSE file is restrictive (Anthropic proprietary terms); confirm you have rights to use/copy the code as packaged. (4) Run these scripts locally in a controlled environment and review their behavior on sample PDFs before processing sensitive documents. If you need an automated install, request/confirm an install spec that explicitly installs and pins the required dependencies rather than relying on manual setup.

Type: OpenClaw Skill Name: pdf-anthropic Version: 1.0.0 The bundle provides a comprehensive and legitimate set of tools for PDF manipulation, including text/table extraction, merging, splitting, and form filling. It utilizes standard Python libraries such as pypdf, pdfplumber, and reportlab, along with clear instructions in SKILL.md and forms.md for an AI agent to execute these tasks. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found; the scripts and documentation are entirely consistent with the stated purpose of advanced PDF processing.