MCP PDF Reader

MCP PDF Reader is an MCP server that provides tools for processing PDF documents through text parsing, image extraction, and optical character recognition. It connects local PDF files to AI assistants using FastMCP, PyMuPDF, Pillow, and the Tesseract OCR engine. Software engineers, researchers, and data analysts use this server to read standard text layers, extract embedded graphical assets, and perform OCR on scanned pages or embedded images within documents. The server enables clients to inspect structural PDF metadata, parse specific page ranges, and support multiple OCR languages such as English, French, German, and Spanish. By exposing dedicated tools for textual and visual extraction, it allows language models to thoroughly inspect documents containing both machine-readable text and non-selectable scanned graphics without relying on third-party cloud conversion services.

Category: Design, Media & Creative

Tags: image-processing, ocr, pdf, tesseract, text extraction

Visit MCP PDF Reader

How to install and configure MCP PDF Reader

1. Install the Tesseract OCR system binary on your operating system (for example, brew install tesseract on macOS or sudo apt install tesseract-ocr on Ubuntu/Debian). 2. Clone the repository and install dependencies using UV: bash git clone https://github.com/labeveryday/mcp_pdf_reader.git cd mcp_pdf_reader uv sync 3. Alternatively, install dependencies in a Python virtual environment: bash pip install fastmcp PyMuPDF pytesseract Pillow 4. Add the server configuration to your claude_desktop_config.json file: json { "mcpServers": { "pdf-reader": { "command": "uv", "args": ["run", "python", "/path/to/your/pdf_reader_server.py"], "cwd": "/path/to/your/mcp-pdf-reader-server" } } } 5. Restart Claude Desktop to activate the tools.

What you can do with MCP PDF Reader

• Extract text content and page-level word counts across selected page ranges within local PDF documents. - Extract embedded images from specific PDF pages and save them to a designated local output directory. - Perform OCR on scanned PDF pages or embedded images using single or multi-language Tesseract profiles. - Retrieve comprehensive document metadata, statistics, and page summaries to inspect PDF file properties. - Analyze PDF document structure and visual content distribution across pages for automated data workflows.

Key facts

• https://github.com/labeveryday/mcp_pdf_reader
• Design, Media & Creative, Files, Documents & PDFs
• image-processing, ocr, pdf, tesseract, text extraction

Part of MCP Servers

What is MCP PDF Reader?

MCP PDF Reader is a Model Context Protocol server built with FastMCP. It enables AI clients like Claude Desktop to parse PDF documents, extract regular text layers, pull embedded images, and perform optical character recognition via Tesseract.

What system dependencies are required for MCP PDF Reader?

The server requires Tesseract OCR installed on the host machine and available in the system PATH. Python dependencies include fastmcp, PyMuPDF, pytesseract, and Pillow, which handle PDF rendering and optical character extraction.

What tools are available in MCP PDF Reader?

The server provides five tools: read_pdf_text for reading standard text, extract_pdf_images for exporting image files, read_pdf_with_ocr for running OCR on pages and images, get_pdf_info for reading metadata, and analyze_pdf_structure for inspecting content layout.

Does MCP PDF Reader support languages other than English?

Yes. It supports multiple languages through Tesseract language packs. Users can pass language codes like fra for French, deu for German, spa for Spanish, or combined strings like eng+fra into the read_pdf_with_ocr tool.

What license does MCP PDF Reader use?

MCP PDF Reader is released under the MIT License, allowing users to inspect, modify, and distribute the software according to the repository license terms.