pdf-reader-mcp-server

the-pujon/pdf-reader-mcp-server

3.1

If you are the rightful owner of pdf-reader-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The PDF Reader MCP Server is a tool that allows AI assistants to interact with PDF documents directly within development environments, providing capabilities such as text extraction, search, and metadata retrieval.

Tools
4
Resources
0
Prompts
0

PDF Reader MCP Server

A Model Context Protocol (MCP) server that enables AI assistants like Cursor to read, search, and analyze PDF documents directly within your development environment.

Features

  • 📄 Full PDF Text Extraction - Access complete document content
  • 🔍 Intelligent Search - Find text with contextual previews
  • 📊 Document Metadata - Get page counts, word counts, and PDF info
  • ✂️ Excerpt Extraction - Extract specific text ranges
  • 🚀 Fast Performance - Loads once, queries instantly
  • 🔌 MCP Integration - Seamless integration with Cursor IDE

Prerequisites

  • Node.js 18 or higher
  • npm or yarn
  • Cursor IDE with MCP support

Installation

  1. Clone or download the project

    git clone <repository-url>
cd pdf-mcp-server

  2. Install dependencies

    npm install

  3. Place your PDF file

    • Put your PDF in the project directory and name it document.pdf, OR
    • Set the PDF_PATH environment variable to point to your PDF file

Configuration

Method 1: Environment Variable

# Windows
set PDF_PATH=C:\path\to\your\document.pdf

# macOS/Linux
export PDF_PATH=/path/to/your/document.pdf

Method 2: Edit the Code

Update the PDF_PATH variable in server.js:

const PDF_PATH = path.join(__dirname, "your-document.pdf");

Usage

Running the Server

npm start
# or
node server.js

Configuring Cursor IDE

Add the server to your Cursor MCP configuration. The location varies by OS:

  • Windows: %APPDATA%\Cursor\User\globalSettings\settings.json
  • macOS: ~/Library/Application Support/Cursor/User/globalSettings/settings.json
  • Linux: ~/.config/Cursor/User/globalSettings/settings.json

Add this configuration:

{
  "mcp": {
    "servers": {
      "pdf-reader": {
        "command": "node",
        "args": ["path/to/your/pdf-mcp-server/server.js"],
        "env": {
          "PDF_PATH": "path/to/your/document.pdf"
        }
      }
    }
  }
}

Available Tools

get_pdf_content

Get the complete text content of the loaded PDF document.

Parameters: None

Example:

Please show me the content of the PDF document.

search_pdf

Search for specific text within the PDF document.

Parameters:

  • query (required): Text to search for
  • case_sensitive (optional): Whether to perform case-sensitive search (default: false)

Example:

Search for "machine learning" in the PDF document.

get_pdf_info

Get metadata and information about the loaded PDF.

Parameters: None

Example:

What information can you tell me about this PDF?

get_pdf_excerpt

Extract a specific portion of the PDF by character range.

Parameters:

  • start (required): Starting character position
  • length (optional): Number of characters to extract (default: 1000)

Example:

Get an excerpt from the PDF starting at character position 1500.

Available Resources

pdf://document

Direct access to the PDF content as a resource that can be referenced by the AI assistant.

Example Interactions

Once configured, you can interact with your PDF through Cursor's AI assistant:

  • "What is this document about?"
  • "Search for any mentions of 'API authentication' in the PDF"
  • "Summarize the section about database design"
  • "Find all code examples in the document"
  • "What are the main topics covered in this PDF?"

Troubleshooting

PDF Not Found Error

PDF file not found at: /path/to/document.pdf

Solution: Ensure the PDF file exists at the specified path or update the PDF_PATH variable.

Server Won't Start

Failed to start server: Error message

Solutions:

  • Check that Node.js version is 18 or higher
  • Ensure all dependencies are installed: npm install
  • Verify the PDF file is readable

Cursor Not Detecting Server

Solutions:

  • Restart Cursor IDE after adding the MCP configuration
  • Check that the server path in the configuration is correct
  • Verify the server starts without errors when run manually

Development

Project Structure

pdf-mcp-server/
├── server.js          # Main MCP server implementation
├── package.json       # Node.js dependencies and scripts
├── README.md          # This file
└── pdf/document.pdf       # Your PDF file (not included)

Dependencies

  • @modelcontextprotocol/sdk - MCP TypeScript SDK
  • pdf-parse - PDF text extraction
  • zod - Schema validation

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Test with different PDF files
  5. Submit a pull request

Future Enhancements

  • Support for multiple PDF files
  • Page-by-page navigation
  • OCR support for scanned PDFs
  • Table extraction
  • Image description
  • Bookmark and outline parsing
  • Password-protected PDF support

License

MIT License - see LICENSE file for details.

Acknowledgments

Need help? Open an issue or reach out with questions about setup and usage.