Docsmith

Provides Python-powered document processing through MCP to read and write Excel, Word, PDF, and text files with optional pagination.

typescript

0

GitHub Stars

typescript

Language

4 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": {
    "mcpc-tech-docsmith-mcp": {
      "command": "node",
      "args": [
        "/path/to/docsmith-mcp/dist/index.js"
      ],
      "env": {
        "DOC_PAGE_SIZE": "100",
        "DOC_MAX_FILE_SIZE": "50",
        "DOC_RAW_FULL_READ": "false"
      }
    }
  }
}

Docsmith MCP is a Python-powered document processing MCP that lets you read, extract, and write content from common document formats such as Excel, Word, PDF, and text files. It runs in a secure WebAssembly sandbox via Pyodide and is designed to integrate with your MCP client to perform document operations efficiently and with pagination support where needed.

How to use

You will integrate Docsmith MCP into your MCP client by listing it as an MCP server and then invoking its tools to read or write documents. Start by ensuring the server is available to your client, then choose the appropriate tool for your task: read_document to extract content with automatic format detection, write_document to create or update files, and get_document_info to retrieve metadata about a document. Use paginated reading for large files to avoid loading everything into memory. You can control how much content is returned per page and whether you want a full raw read or a paginated view.

How to install

Prerequisites: you need Node.js on your machine, and your MCP client should be able to connect to an MCP server configured to run locally or remotely.

Install and build the MCP server package following the runtime instructions provided in the setup flow.

Available tools

read_document

Read document content with automatic format detection. Supports paginated or full raw reading modes depending on the requested page size and mode.

write_document

Write document content to a target file in the specified format, such as Excel or Word, using provided data arrays.

get_document_info

Retrieve metadata for a given document file, including basic properties and structure.