PDF Manipulation Toolkit

What is PDF Manipulation Toolkit?

PDF Manipulation Toolkit is an MCP (Model Context Protocol) server that integrates with Claude AI to perform complex PDF operations using natural language commands. It supports merging, splitting, encrypting, optimizing, analyzing, and repairing PDFs via a set of 16 tools powered by PDFtk and QPDF.

How to use PDF Manipulation Toolkit?

Clone the repository, create a Python virtual environment, install dependencies from requirements.txt, and install external tools PDFtk and QPDF. Then add the server to the Claude Desktop configuration file (claude_desktop_config.json) pointing to server.py. After restarting Claude Desktop, issue natural language commands like “Merge these 3 PDFs into one document” or “Encrypt my report with password protection”. Alternatively, call tools programmatically using their Python signatures.

Key features of PDF Manipulation Toolkit

• Merge, split, extract, and rotate PDF pages
• AES-256 and basic password encryption/decryption
• Optimize PDFs for web or email size
• Repair corrupted PDFs and validate integrity
• Extract embedded attachments and update metadata
• Download PDFs from URLs and preview via system viewer
• Flexible file path resolution with workspace configuration

Use cases of PDF Manipulation Toolkit

• Combine multiple research papers into a single bibliography
• Split a large manual into separate chapter files
• Encrypt a confidential contract with password protection
• Optimize a batch of PDFs for sharing over email
• Fix a corrupted presentation before a deadline

FAQ from PDF Manipulation Toolkit

What dependencies are required?

Python 3.8 or later, PDFtk, and QPDF must be installed on the system. Python packages are listed in requirements.txt (FastMCP, requests, pathlib).

How do I configure the workspace directory?

Use the configure_pdf_workspace tool with an absolute path, or set the PDF_WORKSPACE environment variable. The server searches default directories (~/Documents/PDFs, ~/Downloads, etc.) for filename-only references.

What PDF operations can I perform?

You can merge, split, extract pages, rotate, encrypt (AES‑256 or basic), decrypt, optimize, repair, check integrity, get metadata, update metadata, inspect structure, extract attachments, download, preview, count PDFs in a directory, and manage workspace.

How do I check if tools are installed correctly?

Run get_server_status() which verifies PDFtk and QPDF availability. Use list_default_directories() to confirm search paths.

What should I do if a file is not found?

Check that the file path is correct or that the filename exists in one of the default directories. Use get_file_info to see path details. The server supports absolute, relative, and filename-only paths.