mcp-doc-forge

The cablate-mcp-doc-forge server exposes comprehensive document processing capabilities, including reading various document formats (DOCX, PDF, TXT, HTML, CSV), converting between formats (DOCX to HTML/PDF, HTML to TXT/Markdown), and manipulating PDFs (merge, split). It also offers text processing utilities such as multi-encoding support (UTF-8, Big5, GBK), text cleaning and formatting, diff generation, and text splitting by lines or delimiters. For HTML processing, you can clean, format, and extract resources (images, links, videos) while preserving structure during conversions. To use the server in your MCP-enabled workflow, install the MCP server via npx as shown in the installation section, then connect to it from your MCP client or Dive Desktop by referencing the server as cablate-mcp-doc-forge with the appropriate command configuration. The server is designed to be invoked via MCP tooling, enabling seamless document workflows across your models and clients.

Prerequisites:

• Node.js and npm installed on your system
• Access to the internet to fetch the MCP package

Installation steps:

1. Install using Smithery (automatic)

npx -y @smithery/cli install @cablate/mcp-doc-forge --client claude

If you prefer manual installation or need to run locally:

2. Install the MCP package globally

npm install -g @cablate/mcp-doc-forge

3. Run the MCP server directly (example)

mcp-doc-forge

4. If you are configuring via Dive Desktop or another MCP client, ensure you reference the server with:

{
  "mcpServers": {
    "cablate-mcp-doc-forge": {
      "command": "npx",
      "args": ["-y", "@cablate/mcp-doc-forge"],
      "enabled": true
    }
  }
}

Notes:

• Using npx ensures you run the latest version from the npm registry without a global install.
• If you install locally, ensure your PATH includes the directory where the binary is installed or invoke via npx as shown above.

Tips and common issues:

• Ensure Node.js version is compatible with the mcp-doc-forge package.
• When using Dive Desktop, copy the exact configuration snippet provided for the mcpServers section to avoid connection issues.
• If you encounter network or npm registry access problems, try clearing npm cache or using a different registry mirror.
• The server supports multiple document formats; for best results, keep input documents standardized (e.g., consistent encoding like UTF-8).
• Monitor logs for errors related to specific formats (PDF parsing can require additional native dependencies in some environments).
• Environment variables can be used to customize behavior in deployment; consider adding logging level or output directories if your runtime requires them.