MCP PDF Reader Server (Python + FastMCP)

A powerful Model Context Protocol (MCP) server built with FastMCP that provides comprehensive PDF processing capabilities including text extraction, image extraction, and OCR for reading text within images.

Features

• Text Extraction: Extract text content from PDF pages
• Image Extraction: Extract all images from PDF files
• OCR Capabilities: Read text from images using Tesseract OCR
• Comprehensive Analysis: Get detailed PDF structure and metadata
• Page Range Support: Process specific page ranges
• Multiple Languages: OCR support for multiple languages

Prerequisites

System Dependencies

Tesseract OCR

You need to install Tesseract OCR on your system:

Ubuntu/Debian:

sudo apt update
sudo apt install tesseract-ocr tesseract-ocr-eng

macOS:

brew install tesseract

Windows:

1. Download from: https://github.com/UB-Mannheim/tesseract/wiki
2. Install and add to PATH
3. Or use: conda install -c conda-forge tesseract

Additional Language Packs (Optional)

# For multiple languages
sudo apt install tesseract-ocr-fra tesseract-ocr-deu tesseract-ocr-spa

Installation

Quick Start with UV

1. Install UV (if not already installed):

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

2. Clone/Create the project:

mkdir mcp-pdf-reader-server
cd mcp-pdf-reader-server

3. Initialize and install with UV:

# Copy the files (pdf_reader_server.py and pyproject.toml)
# Then install dependencies
uv sync

4. Verify installation:

uv run python -c "import pytesseract; print(pytesseract.get_tesseract_version())"

Alternative: Manual Setup

If you prefer traditional setup:

1. Create virtual environment:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

2. Install dependencies:

pip install fastmcp PyMuPDF pytesseract Pillow

Usage

Running the Server

With UV:

uv run python pdf_reader_server.py

Or if you have the environment activated:

python pdf_reader_server.py

The server will start and listen for MCP requests on stdin/stdout.

Available Tools

1. read_pdf_text

Extract text content from PDF pages.

Parameters:

• file_path (string, required): Path to the PDF file
• page_range (object, optional): Dict with start and end page numbers

Example:

{
  "file_path": "/path/to/document.pdf",
  "page_range": {"start": 1, "end": 5}
}

2. extract_pdf_images

Extract all images from a PDF file.

Parameters:

• file_path (string, required): Path to the PDF file
• output_dir (string, optional): Directory to save images
• page_range (object, optional): Page range to process

Example:

{
  "file_path": "/path/to/document.pdf",