---

name: pdf description: > PDF processing toolkit. TRIGGER when the user needs to: merge, split, extract text/tables/images from, create, OCR, watermark, encrypt/decrypt, rotate, or programmatically manipulate PDF files. Also trigger for converting PDFs to other formats or analyzing PDF content at scale. DO NOT trigger for reading a single PDF file (use the Read tool directly).

PDF Processing Guide

Python Libraries

• pypdf: merge (PdfWriter.add_page()), split, rotate, metadata, encrypt (writer.encrypt()), watermark (page.merge_page())
• pdfplumber: text extraction (page.extract_text()), table extraction (page.extract_tables() → pandas)
• reportlab: create PDFs (Canvas for simple, Platypus/SimpleDocTemplate for multi-page)
• pytesseract + pdf2image: OCR scanned PDFs (convert_from_path() → image_to_string())

CLI Tools

• pdftotext (poppler): text extraction, -layout preserves layout
• qpdf: merge (qpdf --empty --pages), split, rotate, decrypt
• pdftk: merge, split (burst), rotate

Quick Reference

Task	Best Tool	Key API
Extract text	pdfplumber	page.extract_text()
Extract tables	pdfplumber	page.extract_tables() → DataFrame
Merge	pypdf	PdfWriter().add_page()
Split	pypdf	iterate reader.pages
Create	reportlab	SimpleDocTemplate / Canvas
OCR	pytesseract	convert_from_path() → image_to_string()
Watermark	pypdf	page.merge_page()
Password	pypdf	writer.encrypt()

User Request

$ARGUMENTS