Skip to content
MCP Shelf

PDF skill for Claude

Reviewed by Sophie Filer · Updated 27 Oct 2026

Reads, splits, merges, fills and builds PDF files with proven Python and command line tools.

This is one of the document skills behind file handling on claude.ai, and it is just as useful in Claude Code. When a task involves a PDF, Claude reaches for it and picks the right tool for the job rather than improvising.

It covers the everyday jobs (pull text or tables out, join or split files, rotate pages, add a watermark) as well as the fiddly ones: filling in form fields and running OCR on scanned pages.

When Claude reaches for it

  • Pulling text or tables out of a PDF
  • Merging, splitting or rotating PDF files
  • Filling a PDF form, or making a scan searchable
  • Producing a new PDF report

What is in the pod

  • A quick reference that maps each job to a Python library or command line tool
  • A separate, step-by-step guide for filling in forms
  • Helper scripts for form fields and page checks

Planting it

Claude Code

  1. Add the Anthropic skills marketplace once.
  2. Install the document-skills plugin, which includes pdf.
  3. Start a new session. Claude loads the skill on its own when a task matches.
Claude Code
/plugin marketplace add anthropics/skills
/plugin install document-skills@anthropic-agent-skills

Claude apps (web, desktop, mobile)

  1. On a paid plan, turn on code execution and file creation in Settings.
  2. The document skills are built in, so there is nothing else to add.

What it needs

  • Python 3 with pypdf, pdfplumber and reportlab
  • poppler and qpdf for the command line tools
  • Tesseract for OCR on scanned pages

Notes from the pea patch

  • For tables, ask Claude to show you the first extracted table before it processes the rest. Layouts vary and one check saves a redo.
  • On claude.ai this works out of the box with file creation turned on; in Claude Code you need the Python packages above.

Pea-fect partners on the shelf