Excel MCP Server

What is Excel MCP Server?

Excel MCP Server is a Model Context Protocol (MCP) server that lets you manipulate Excel files without needing Microsoft Excel installed. It enables AI agents to create, read, and modify Excel workbooks.

How to use Excel MCP Server?

Install Python 3.10 or higher, clone the repository, and install with uv pip install -e .. Start the server with uv run excel-mcp-server (default port 8000) or set the FASTMCP_PORT environment variable for a custom port. Configure your AI tool (e.g., Cursor IDE) to connect via SSE at the server’s URL, and optionally set the EXCEL_FILES_PATH environment variable.

Key features of Excel MCP Server

• Create and modify Excel workbooks
• Read and write data
• Apply formatting and styles
• Create charts and visualizations
• Generate pivot tables
• Manage worksheets and ranges

Use cases of Excel MCP Server

• Let an AI assistant generate and edit Excel reports
• Automate creation of charts and pivot tables from data
• Manage multiple worksheets and ranges programmatically
• Batch update or format existing Excel files
• Integrate Excel manipulation into agent workflows without Excel installed

FAQ from Excel MCP Server

What are the prerequisites for using Excel MCP Server?

Python 3.10 or higher is required. No Microsoft Excel installation is needed.

What transport protocol does Excel MCP Server use?

It uses Server-Sent Events (SSE). For use with Claude Desktop (which requires stdio), you can use Supergateway to convert SSE to stdio.

How do I configure Excel MCP Server with Cursor IDE?

Add a JSON configuration to Cursor with the server’s SSE URL (e.g., http://localhost:8000/sse) and set the EXCEL_FILES_PATH environment variable to the directory containing your Excel files.

Where are Excel files stored by default?

They are stored in the ./excel_files directory relative to where the server runs. This path can be changed with the EXCEL_FILES_PATH environment variable.