---
title: "PDF Skill"
type: "skill"
provider: "anthropic"
category: "tools"
source: "https://github.com/anthropics/skills"
license: "MIT"
tags: ["pdf", "documents", "skill", "tools"]
dateAdded: "2026-03-19"
featured: true
origin: "official"
format: "skill-md"
description: "A reusable Skill that teaches an agent to read, fill, merge, split and OCR PDF files. Skills package instructions plus scripts an agent can invoke."
---

# PDF Skill

Use this skill whenever the user wants to do anything with PDF files: reading or
extracting text/tables, merging or splitting documents, rotating pages, adding
watermarks, filling forms, encrypting/decrypting, extracting images, or running OCR
on scanned PDFs to make them searchable.

## How to use

1. Identify the operation the user wants (extract, merge, split, fill, ocr, …).
2. Call the matching script with the input path(s) and desired output path.
3. Report the result path and a one-line summary of what changed.

## Do not use for

- Word documents (.docx) — use the docx skill.
- Spreadsheets — use the xlsx skill.
- General coding tasks unrelated to documents.