Installation

“PC Load Letter? What the f** does that mean?”*

Don’t worry. This is simpler than fixing the printer.

Quick Install

# Run directly without installing
uvx mcwaddams

This downloads and runs mcwaddams in an isolated environment. No global installs, no dependency conflicts.

pip install mcwaddams

uv add mcwaddams

claude mcp add mcwaddams "uvx mcwaddams"

That’s it. The server will be available in your next Claude Code session.

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "mcwaddams": {
      "command": "uvx",
      "args": ["mcwaddams"]
    }
  }
}

mcwaddams is a standard MCP server. Any client that speaks the Model Context Protocol can use it:

# Generic stdio transport
uvx mcwaddams

Start your MCP client (Claude Code, Claude Desktop, etc.)
Check available tools

Ask: “What Office document tools do you have?”

You should see 20 tools including extract_text, convert_to_markdown, analyze_excel_data, etc.
Test extraction

Point it at any Office document:
```
Extract text from /path/to/document.docx
```

mcwaddams bundles everything it needs:

Library	Purpose
`python-docx`	Modern Word documents
`openpyxl`	Modern Excel spreadsheets
`python-pptx`	Modern PowerPoint
`mammoth`	Word to Markdown fallback
`pandas`	Excel fallback + CSV
`olefile`	Legacy OLE formats (.doc, .xls, .ppt)
`xlrd`	Legacy Excel support
`Pillow`	Image extraction

Make sure you’re using Python 3.11 or higher:

python --version  # Should be 3.11+

Run your terminal as Administrator, or check that Python has access to the document locations.

Next: Quick Start — Extract your first document

🎉

Flair Earned!

Badge Name