PDF Reader

PDF Reader MCP Server for VS Code

python

6

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "pietermyb-mcp-pdf-reader": {
      "command": "uvx",
      "args": [
        "pdf-reader-mcp"
      ]
    }
  }
}

This PDF Reader MCP Server lets VS Code connect to and analyze PDF documents through the Model Context Protocol (MCP). You can open PDFs, extract text and metadata, generate summaries, and answer questions about content directly from your editor or other MCP-compatible clients.

How to use

Install, run, and connect your MCP client to the server to start analyzing PDFs. Use the provided tools to open a PDF, view metadata, count pages, extract text from pages or entire documents, and generate summaries. You can also ask questions about the content and have the server analyze specific page ranges.

How to install

Prerequisites: Python 3.13 or newer.

Install the MCP server using a local development setup or Docker.

uv venv .venv
source .venv/bin/activate

uv pip install -e .

# Optional: build and run locally for development
uv sync
uv build
uv publish

Configuration and running the MCP server

You can run the MCP server locally using the standard MCP runtime or via the Docker image. Use the following standard runtime configuration for an MCP-enabled client to connect.

{ 
  "mcpServers": [ 
    {
      "type": "stdio",
      "name": "pdf_reader",
      "command": "uvx",
      "args": ["pdf-reader-mcp"]
    }
  ],
  "envVars": []
}

Notes on wiring with clients

Choose the appropriate client configuration to start communicating with the MCP server. If you are using VS Code, ensure the MCP extension is installed and the client is set to run the stdio-based server command with the provided arguments. If you are using Claude Desktop or another MCP-compatible client, configure the mcpServers section accordingly to launch the server using the same stdio approach.

Examples of common workflows

Open a PDF, fetch its metadata, and count pages to understand the document scope. Then extract text from all pages or a specific range, and finally generate a concise summary of the document. You can also prompt the server to answer questions about the content or extract text from particular pages for targeted analysis.

Development and testing tips

Develop against the server locally by running the stdio configuration and testing through your MCP client. You can test tools like open-pdf, list-pdf-metadata, get-pdf-page-count, get-pdf-page-text, pdf-to-text, and the prompts summarize-pdf, extract-text-from-pdf, and analyze-pdf. Use a virtual environment to isolate dependencies and validate changes with your editor’s MCP client.

Available tools

open-pdf

Open a PDF file and obtain a unique PDF ID for subsequent operations.

close-pdf

Close an open PDF using its PDF ID.

list-pdf-metadata

View metadata for an open PDF identified by its PDF ID.

get-pdf-page-count

Return the total number of pages in the PDF identified by its PDF ID.

get-pdf-page-text

Return the text content of a specific page in a PDF by page number.

pdf-to-text

Extract all text from a PDF with optional range and page-number inclusion.

summarize-pdf

Generate a summary of a PDF with optional detail level.

extract-text-from-pdf

Extract text from specific pages or page ranges of a PDF.

analyze-pdf

Analyze a PDF and answer questions about its content, optionally focusing on a page range.