PDF Skill for Claude: Read, Merge, Fill and Create PDFs

/plugin install document-skills@anthropic-agent-skills

The pdf skill is one of Anthropic’s document skills, the same family that powers Claude’s file creation features. It tells Claude how to handle almost any PDF task with Python libraries such as pypdf, pdfplumber and reportlab.

What it does

According to its description, the skill covers:

• Reading and extracting text and tables.
• Merging several PDFs, splitting them apart, rotating pages.
• Adding watermarks, creating new PDFs.
• Filling PDF forms (with a dedicated forms.md reference).
• Encrypting and decrypting, extracting images, and OCR on scanned PDFs.

Install

The four document skills (pdf, docx, xlsx, pptx) ship together:

/plugin marketplace add anthropics/skills
/plugin install document-skills@anthropic-agent-skills

On claude.ai, these capabilities are already available on paid plans. In Claude Code, the needed Python packages must be installed in your environment. See the skills guide for how skills load.

Example usage

Use the PDF skill to extract the form fields from path/to/application.pdf
Merge every invoice in ./invoices into one PDF sorted by date
Pull the tables from report.pdf into a CSV
Add a "DRAFT" watermark to every page of proposal.pdf

Tips

• Name the file path explicitly so Claude knows which document to open.
• For large PDFs, ask for extraction to a file rather than printing everything into the conversation.
• Need the output as a spreadsheet or Word file? Combine with xlsx or docx.

Source and license

Source: anthropics/skills/skills/pdf. License: proprietary, source-available; see LICENSE.txt. Anthropic shares it as a reference for complex, production-grade skills.