by Anthropic · anthropics/skills
Comprehensive PDF manipulation: extract text & tables, create PDFs, merge/split, fill forms, OCR scanned documents.
89K
Installs
75K
GitHub Stars
4.9
Rating
4
Platforms
Try in Playground
No installation needed — experience this skill directly in your browser.
Example input:
Extract all tables from this quarterly report PDF and summarize the key financial figures.
Skill Instructions
This is the system prompt loaded when the skill is activated (core SKILL.md content):
You are a PDF processing expert. You can: 1. Extract text and tables from PDFs with high fidelity 2. Generate new PDFs from structured content 3. Merge multiple PDFs into one 4. Split a PDF into separate pages 5. Fill PDF form fields 6. OCR scanned PDFs to extract text from images When given a PDF task, use the appropriate tool (pdfplumber, reportlab, PyPDF2, pytesseract) and produce clean, well-formatted output. Always preserve document structure and report any extraction issues.
Tags
Install Command
/plugin install document-skills@anthropic-agent-skillsCompatible With
Permissions & Security
Safe
Static scan passed
Read/write PDF files in workspace
Always review source code before installing, especially skills requesting filesystem or network access.
In These Packs
Links
GitHub RepositoryRelated Skills
The #1 installed meta-skill — lets your agent search and discover other skills from the ecosystem conversationally.
Create and edit Word documents with tracked changes, comments, styles, tables, images, and formatting preservation.
Automate Gmail, Sheets, Drive, and Calendar — read inquiries, extract data, schedule follow-ups, archive documents.
Interactive meta-skill that guides you through creating new SKILL.md files with proper structure, scripts, and examples.