PDF MCP Server

Extract text and data from PDF files using pdfplumber by providing local file paths.

PDF MCP Server

A simple Model Context Protocol (MCP) server that provides PDF reading capabilities using pdfplumber.

Features

  • extract_text: Extract all text content from PDF files
  • get_pdf_info: Get basic PDF information (page count, metadata, dimensions)
  • extract_tables: Extract tables from PDF files

Installation

Option 1: Without installation (using uvx)

You can run the server directly without installing it:

uvx --from git+https://github.com/gstiebler/pdf-mcp-server.git pdf-mcp-server

Option 2: Local installation

Clone the repository and install dependencies:

git clone https://github.com/gstiebler/pdf-mcp-server.git
cd pdf-mcp-server
uv sync

Usage

If installed locally:

python main.py

The server runs using stdio transport and can be integrated with any MCP client.

Available Tools

extract_text

Extracts all text content from a PDF file.

Parameters:

  • file_path (string): Path to the PDF file

Returns:

  • Complete text content from all pages

get_pdf_info

Gets basic information about a PDF file.

Parameters:

  • file_path (string): Path to the PDF file

Returns:

  • Dictionary containing:
    • page_count: Number of pages
    • metadata: PDF metadata (if available)
    • page_dimensions: Width and height of the first page

extract_tables

Extracts tables from a PDF file.

Parameters:

  • file_path (string): Path to the PDF file
  • page_number (integer, optional): Specific page number to extract from (1-indexed)

Returns:

  • List of tables with:
    • page: Page number where the table was found
    • data: Table data as a list of rows (each row is a list of cell values)

Integration with Claude Desktop

To use this server with Claude Desktop, add it to your Claude Desktop configuration:

Option 1: Direct from GitHub (recommended)

{
  "mcpServers": {
    "pdf-reader": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/gstiebler/pdf-mcp-server.git", "pdf-mcp-server"]
    }
  }
}

Option 2: From local installation

{
  "mcpServers": {
    "pdf-reader": {
      "command": "python",
      "args": ["/path/to/pdf-mcp-server/main.py"]
    }
  }
}

Requirements

  • Python 3.10+
  • pdfplumber
  • mcp

Máy chủ liên quan

NotebookLM Web Importer

Nhập trang web và video YouTube vào NotebookLM chỉ với một cú nhấp. Được tin dùng bởi hơn 200.000 người dùng.

Cài đặt tiện ích Chrome