---

name: pdf category: integration-connectivity/mcp-integrations version: 1.0.0 description: Extract text, tables, metadata from PDFs with merge and annotation support author: blackbox5/mcp verified: true tags: [mcp, pdf, documents, processing, extraction]

PDF Processing

When to Use:

• Processing contracts and legal documents
• Extracting data from reports
• Combining multiple PDFs into one
• Adding annotations or watermarks
• Converting PDFs to other formats

Always handle errors gracefully and validate PDF structure before processing.

<available_skills> <skill_group name="Text Extraction"> Extract all text from PDF Extract text from document.pdf Extract text from specific pages Get text from pages 1, 3, and 5 List of page numbers to extract (1-indexed) </skill_group>

<skill_group name="Table Extraction"> Extract all tables from PDF Pull all tables from report.pdf Save extracted tables to CSV Convert tables to output.csv </skill_group>

<skill_group name="Metadata Operations"> Extract PDF metadata Get title, author, creation date Title, author, subject, creator, producer, creation date, page count </skill_group>

<skill_group name="PDF Manipulation"> Combine multiple PDFs into one Merge file1.pdf, file2.pdf, file3.pdf into merged.pdf Split PDF into individual pages Split document.pdf into separate page files Rotate pages by specified degrees Rotate all pages 90 degrees clockwise Merge specific pages from multiple PDFs Merge pages 1-3 from file1 and pages 5-7 from file2 </skill_group>

<skill_group name="Annotation"> Add text watermark to all pages Add 'CONFIDENTIAL' watermark to each page </skill_group>

<skill_group name="OCR"> Extract text from scanned PDF using OCR Extract text from scanned.pdf with OCR Requires Tesseract to be installed </skill_group> </available_skills>

<best_practices> Validate PDF before processing Handle encoding issues properly Use error handling for file operations Test on small samples first Process pages in chunks for large PDFs Backup original PDFs before modification Assume all PDFs are text-based Process entire large PDF at once Ignore page ranges for extraction Forget to close file handles Hardcode file paths </best_practices>

<error_handling> Invalid PDF format Verify file is valid PDF Check file extension is .pdf Try opening in PDF viewer Encrypted PDF Check if PDF is password protected Provide password if available Use appropriate decryption tool No text found PDF might be scanned/image-based Use OCR for extraction Check for embedded fonts Table extraction fails Try different extraction method Check table structure complexity Manually verify table boundaries OCR not working Verify Tesseract is installed Check image quality of pages Try preprocessing images </error_handling>

<integration_notes> Basic text extraction, merging, splitting Advanced table extraction OCR for scanned PDFs Adding watermarks and annotations </integration_notes>

<advanced_operations> Batch Processing Process multiple PDFs in parallel Form Data Extraction Extract data from PDF forms Image Extraction Extract embedded images from PDF Compression Reduce PDF file size </advanced_operations>