Anthropic PDF Tools Skill

What it is: Anthropic's reference PDF Tools Skill, distributed in the official anthropics/skills repository. It teaches Claude how to read, extract text, query, and manipulate PDF documents from inside an agent session.

Why it matters for data work

So much business-critical data lives in PDFs — invoices, financial filings, scientific papers, government datasets. This skill turns a folder of PDFs into a queryable surface that an agent can reason over without you wiring up a custom pipeline first.

Install & configure

Clone the official skills repo and copy the PDF skill into your local Claude Code skills directory:

git clone https://github.com/anthropics/skills
cp -r skills/document-skills/pdf ~/.claude/skills/

Once installed, Claude Code will auto-discover the skill via its SKILL.md frontmatter and invoke it whenever a user asks a PDF-related question.

Example usage

Drop a PDF into your project and ask Claude something like "summarize the key financial tables in Q3-report.pdf and write a CSV of the line items". The skill handles extraction; Claude handles the analysis.

Author & links

Author: Anthropic

Repo: github.com/anthropics/skills

License: MIT

Related skills

The same repo ships document skills for DOCX, XLSX, and PPTX — install them together for full Office/PDF coverage in any data-extraction workflow.

← Back to Claude Code Skills