PDF Tools

You can run a PDF Tools MCP Server that exposes common PDF operations like merging files, extracting pages, and searching PDFs. This server enables large language models to manipulate PDFs programmatically, streamlining workflows that involve document assembly, page selection, and content discovery.

How to use

To use the PDF Tools MCP Server with an MCP client, start the local server and point your client to the server command. The server provides these core capabilities: merging multiple PDFs into a single document (including custom-order merging), extracting specific pages from a PDF, and performing searches across PDFs to find relevant content. You can also identify and merge related PDFs based on text extraction and regex patterns from a target input PDF.

How to install

Prerequisites you need before installing:

• Git
• Python installed on your system
• A command-line environment that can run Python commands
• The ability to run the UV runtime (see the installation steps below)

1) Clone the project

git clone <repository-url-for-mcp-pdf-tools>

2) Change to the project directory

cd mcp-pdf-tools

3) Create and activate a virtual environment (Windows example)

uv venv .venv\Scripts\activate

4) Install the package in editable mode

uv pip install -e .

5) Run the server via the UV runtime using the MCP command shown in the config snippet

uv run pdf-tools

Note: The command above expects you to provide a path to your local project in the usage configuration if your environment requires it. You should ensure the environment is set up correctly and that the current directory contains the installed package.

Usage with client configuration

{
  "mcpServers": {
    "pdf_tools": {
      "command": "uv",
      "args": [
        "--directory",
        "PATH_TO\\mcp-pdf-tools",
        "run",
        "pdf-tools"
      ]
    }
  }
}

Available tools

merge_pdfs

Merges multiple PDF files into a single PDF, with the option to specify the order of input files.

extract_pages

Extracts specified pages from a PDF and outputs a new PDF containing only the selected pages.

search_pdfs

Performs a filesystem or content search across PDFs to locate matches for a given query.

find_related_pdfs

Finds and merges related PDFs based on text extraction and regex patterns from a target input PDF.