pdf_skill

Overview

This skill provides a practical toolkit for PDF processing using Python and common CLI tools. It covers extracting text and tables, creating and composing PDFs, splitting and merging files, rotating and watermarking pages, and programmatic form filling. The goal is to streamline document automation, analysis, and batch operations with clear, ready-to-use patterns. Suitable for both quick scripts and production workflows.

How this skill works

The skill uses lightweight Python libraries—pypdf for reading/writing, pdfplumber for layout-aware text and table extraction, reportlab for PDF generation, and pytesseract/pdf2image for OCR on scanned documents. It also demonstrates command-line utilities (pdftotext, qpdf, pdfimages) for fast, scriptable operations. Example snippets show merging, splitting, metadata handling, rotating, watermarking, extracting images/tables, encrypting, and form filling patterns so you can integrate them into apps or pipelines.

When to use it

• Batch converting many PDFs to text or Excel-ready tables for analysis
• Generating reports or multi-page PDFs programmatically from templates
• Merging, splitting, rotating, or watermarking documents before distribution
• Extracting data from tables or images embedded in PDFs for ingestion
• Filling PDF forms automatically in a workflow or application

Best practices

• Prefer pdfplumber for layout-preserving text and table extraction, but fall back to OCR for scanned pages
• Use pypdf for fast in-memory merge/split/rotate operations and to preserve metadata when possible
• Validate extracted tables programmatically (headers, row counts) before downstream use
• Use reportlab or Platypus for templated, repeatable PDF generation rather than drawing text at fixed coordinates
• When handling many files, stream pages rather than loading all PDFs into memory to reduce peak RAM

Example use cases

• Automate monthly report assembly by merging generated pages and stamping a watermark
• Extract invoice line items from PDFs into Excel using pdfplumber + pandas
• Split a large scanned contract into per-signature pages and run OCR to make it searchable
• Fill customer onboarding PDF forms programmatically and produce password-protected copies
• Batch-extract embedded images for archival or image-based analysis using pdfimages