---

name: "document-processor-pdf" description: "Parse, analyze, and extract content from PDF documents with OCR, table detection, and text extraction (Official Anthropic skill)" version: "1.0.0" author: "Anthropic (Adapted for CemaPay)" category: "documents" tags: ["pdf", "document-analysis", "ocr", "text-extraction", "invoice-processing"] requiresAuth: true source: "anthropic-official"

PDF Document Processor Skill

🎯 PURPOSE

Official Anthropic skill for PDF document analysis and extraction. Supports:

• Text Extraction: Pull all text from PDFs (including scanned documents with OCR)
• Table Detection: Parse structured tables and convert to JSON/CSV
• Invoice Processing: Extract invoice data (amounts, dates, line items)
• Form Parsing: Extract form field values
• Image Extraction: Extract embedded images from PDFs

🔍 WHEN TO USE THIS SKILL

Trigger when users mention:

• "extract PDF", "read PDF", "parse document"
• "invoice processing", "extract invoice data"
• "OCR", "scan document", "digitize"
• "table extraction", "form data"

📋 CAPABILITIES

1. Text Extraction

CopyextractPDF({
  file: "invoice.pdf",
  options: {
    ocr: true, // Enable OCR for scanned docs
    preserveLayout: true,
    extractTables: true,
    extractImages: false
  }
});

// Returns
{
  text: "Invoice #12345...",
  tables: [...],
  metadata: { pages: 3, author: "...", created: "..." }
}

2. Invoice Processing

Automatically extract:

• Invoice number
• Invoice date
• Due date
• Vendor/supplier info
• Line items (description, quantity, price)
• Subtotal, tax, total
• Payment terms

3. OCR Support

• Handles scanned documents
• Multiple languages (English, French)
• Handwriting recognition (limited)
• Image quality enhancement

💡 INTEGRATION EXAMPLE

User: "Extract data from this invoice PDF"

AI Response:

CopyI've analyzed the invoice PDF. Here's what I found:

INVOICE DETAILS:
Invoice #: INV-2025-1024
Date: October 24, 2025
Due Date: November 24, 2025

VENDOR:
ABC Supplies Ltd
Douala, Cameroon
Tax ID: CM123456789

LINE ITEMS:
1. Office Supplies - 450,000 XAF
2. IT Equipment - 1,200,000 XAF
3. Furniture - 800,000 XAF

SUBTOTAL: 2,450,000 XAF
VAT (18%): 441,000 XAF
TOTAL: 2,891,000 XAF

Payment Terms: Net 30
Bank Details: [Extracted]

Would you like me to:
1. Save this to accounting system?
2. Create payment transaction?
3. Export to Excel?

🔧 TECHNICAL DETAILS

• Library: pdf-parse (Node.js), pdfplumber (Python)
• OCR Engine: Tesseract OCR
• Supported Formats: .pdf (including PDF/A)
• Max File Size: 50 MB
• Languages: English, French, German, Spanish

📚 DOCUMENTATION

Official Anthropic Documentation: https://docs.anthropic.com/en/docs/build-with-claude/agent-skills

---

Version: 1.0.0
Source: Anthropic Official Skills Library
Last Updated: October 24, 2025