Claude for Excel: The MCP Server You Should Add to Your Claude Code Setup

What It Does

Anthropic recently hosted a webinar for a new tool called Claude for Excel, attended by 10,000 people. This isn't a standalone app; it's a Model Context Protocol (MCP) server designed to integrate directly with Claude Code. It gives Claude the ability to read from, write to, and manipulate Excel and Google Sheets files from your terminal.

This means you can now ask Claude Code to perform complex spreadsheet operations—like merging datasets, cleaning columns, generating reports, or applying formulas—without ever opening Excel. Claude executes these tasks by communicating with the MCP server, which handles the actual file I/O.

Setup

Installation is straightforward via the Claude Code CLI. If you have Claude Code installed, you can add the server to your configuration.

1. Install the server. The exact package name may vary, but MCP servers are typically installed via npm or pip. For example:

   npm install -g @anthropic-ai/claude-for-excel-mcp
   

2. Configure Claude Code. Add the server to your claude_code_config.json (usually in ~/.config/claude-code/):

   {
     "mcpServers": {
       "claude-for-excel": {
         "command": "npx",
         "args": ["-y", "@anthropic-ai/claude-for-excel-mcp"]
       }
     }
   }
   

3. Restart Claude Code. The next time you start a session, Claude will have access to your spreadsheets.

When To Use It

This tool transforms Claude Code from a pure coding assistant into a data workflow automator. Here are specific use cases where it shines:

• Data Pipeline Scripting: Instead of writing a Python script with pandas to clean a CSV, you can tell Claude Code: "Open sales_data.xlsx, remove all rows where the 'Status' column is 'Cancelled', and save the result as cleaned_sales.csv."
• Report Generation: Combine data from multiple sources. Prompt: "Read the customer IDs from sheet1 of this Google Sheet (provide URL), join it with the transaction totals in local_data.xlsx, and create a new spreadsheet with a summary pivot table."
• Formula Debugging & Application: Have Claude audit or apply complex formulas across large sheets. "Check all VLOOKUP formulas in financial_model.xlsx for #N/A errors and correct the table array references."
• Rapid Data Exploration: Quickly profile a dataset. "Open survey_results.csv, give me the count of unique values for each column, and identify any columns with more than 30% missing data."

The key advantage is contextual continuity. You can have Claude write a script to fetch data from an API, then immediately pass that output to the Excel MCP server to format it into a report—all within a single Claude Code session, without switching contexts between your IDE, terminal, and spreadsheet software.