PDF Reader MCP Server

pdf-reader MCP server

An MCP server for reading PDFs

Components

Resources

The server provides academic-aware PDF resources with:

• Custom file:// URI scheme for accessing individual PDFs
• Academic structure detection and key section extraction
• Metadata enriched with document type classification
• Resources optimized for agent understanding

Academic Prompts

The server provides specialized academic analysis prompts:

• summarize-academic-paper: Intelligent academic paper summarization
  • Required "file_path" argument for PDF location
  • Optional "focus" argument (general/methodology/results/implications)
  • Generates prompts with key sections, citations, and metadata
• analyze-research-methodology: Deep methodology analysis
  • Required "file_path" argument for PDF location
  • Focuses on research design, data collection, and statistical methods

Enhanced Tools

Basic PDF Processing:

• load-pdf: Load and cache a PDF file for processing
• get-metadata: Get PDF metadata and document information
• extract-images: Extract embedded images with metadata
• render-page: Render PDF pages as high-resolution images

Academic Enhancements:

• extract-academic-text: Text extraction with proper reading order and math formula preservation
• detect-sections: Identify academic sections (Abstract, Introduction, Methods, Results, etc.)
• extract-abstract: Specifically extract the abstract section
• extract-key-sections: Get key sections optimized for agent understanding
• extract-citations: Parse in-text citations and reference lists
• chunk-content: Break content into agent-friendly semantic chunks
• analyze-document-structure: Comprehensive academic document analysis

Configuration

This PDF reader MCP server provides comprehensive PDF processing capabilities including:

• Full text extraction from any PDF
• High-resolution image extraction
• Table detection and extraction
• Annotation and comment extraction
• PDF metadata retrieval
• Page rendering to images
• Document structure analysis

Installation & Setup

Prerequisites

• Python 3.13 or higher
• uv package manager (install with pip install uv)

Install Dependencies

IDE Integration

VSCode with MCP Extension

1. Install the MCP VSCode Extension
2. Open your VSCode settings (.vscode/settings.json) and add:

WindSurf IDE

1. Open WindSurf settings
2. Navigate to Extensions → MCP Servers
3. Add a new server configuration:

Cursor IDE

1. Open Cursor settings (Cmd/Ctrl + ,)
2. Search for "MCP" or navigate to Extensions → MCP
3. Add server configuration:

Claude Desktop

On MacOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

Starting the Server

Before using the PDF reader in any IDE, start the HTTP server:

The server will start on http://localhost:8000 with the MCP SSE endpoint available at /sse for all IDEs to connect to.

Usage Examples

Once configured in your IDE, you can use the PDF reader with natural language commands:

Basic PDF Processing

Advanced Analysis

Visual Processing

Available Tools

Development

Building and Publishing

To prepare the package for distribution:

1. Sync dependencies and update lockfile:

2. Build package distributions:

This will create source and wheel distributions in the dist/ directory.

3. Publish to PyPI:

Note: You'll need to set PyPI credentials via environment variables or command flags:

• Token: --token or UV_PUBLISH_TOKEN
• Or username/password: --username/UV_PUBLISH_USERNAME and --password/UV_PUBLISH_PASSWORD

Debugging

Since MCP servers run over stdio, debugging can be challenging. For the best debugging experience, we strongly recommend using the MCP Inspector.

You can launch the MCP Inspector via npm with this command:

Upon launching, the Inspector will display a URL that you can access in your browser to begin debugging.