#What it does

The PDF skill handles all PDF operations -- reading, creating, editing, merging, splitting, rotating, watermarking, encrypting/decrypting, extracting images, OCR on scanned documents, and filling forms. It covers both Python libraries and command-line tools.

#How to use

Activate whenever a .pdf file is mentioned or needs to be produced. This includes any PDF manipulation task.

Extract all tables from this PDF into an Excel file

Merge these three PDFs and add a watermark

#Skill instructions

#Tool Selection

| Task | Best Tool | |------|-----------| | Merge PDFs | pypdf | | Split PDFs | pypdf | | Extract text | pdfplumber | | Extract tables | pdfplumber | | Create PDFs | reportlab | | Command-line merge | qpdf | | OCR scanned PDFs | pytesseract | | Fill PDF forms | pdf-lib or pypdf |

#Key Libraries

• pypdf: Basic operations (merge, split, rotate, encrypt, watermark, metadata)
• pdfplumber: Text and table extraction with layout preservation
• reportlab: Creating new PDFs with Canvas or Platypus (multi-page)
• pytesseract + pdf2image: OCR for scanned documents

#Command-Line Tools

• pdftotext: Text extraction with layout preservation
• qpdf: Merge, split, rotate, decrypt
• pdftk: Alternative for merge, split, rotate operations

#Important Notes

• Never use Unicode subscript/superscript characters in ReportLab PDFs (they render as black boxes)
• Use ReportLab's <sub> and <super> XML tags in Paragraph objects instead
• For form filling, consult the dedicated FORMS.md reference

This skill is from the Anthropic Skills Repository.