home / mcp / awels pdf mcp server

Awels PDF MCP Server

A repository with specific Awels' MCP servers

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "awels-io-mcp": {
      "command": "python",
      "args": [
        "-m",
        "src.awels_mcp.run_server"
      ]
    }
  }
}

You have a Python-based MCP server that provides PDF processing capabilities, converting PDFs to Markdown with optional image extraction. This server is designed to run in isolated environments to avoid permission issues, and it exposes a single tool that performs the PDF-to-Markdown conversion for you.

How to use

Start the MCP server locally in a separate terminal, then use an MCP client to invoke the PDF processing tool. The tool can search a directory for PDF files, convert them to Markdown, optionally extract images, and return structured results that include file metadata and processing statistics.

How to install

Prerequisites you need before installing:
- Python 3.10 or newer
- The UVX tool for isolated execution
- Git access to clone sources or install dependencies via pip

Step-by-step installation flow: 1. Create a virtual environment and activate it: 2. Install the package in editable mode using the UV tool: 3. Run tests to verify the installation

uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -e .

Start the MCP server

To start the server, run the following command in a terminal. It launches the MCP server so you can send processing requests from your MCP client.

python -m src.awels_mcp.run_server

Available tools

convert_pdf

Converts PDF files in a directory to Markdown with optional image extraction. Supports recursive search, saving Markdown outputs, and extracting page images or figures.