Best for
- Data analysts automating repetitive Excel workflows
- Developers building Excel-based data solutions
- Business users managing complex Excel workbooks
- Teams maintaining Power Query/VBA/DAX code in Git
M
MCP Server for Excel
Community
AI-driven Excel automation on Windows, powered by the real Excel COM engine.
GitHub source repository ↗
60FMRS · C
A feature-rich, community-maintained MCP server for Windows-only Excel automation that drives the real Excel COM engine to cover Power Query, DAX, VBA, PivotTables, and charts, but it needs a local Excel install and exclusive workbook access, ruling out cross-platform or server-side use.
MCP Server for Excel (repo mcp-server-excel) is an open-source (MIT-licensed) Model Context Protocol server written in .NET that drives the actual Excel application through its Windows COM API, rather than just parsing .xlsx files. It lets conversational AI assistants such as GitHub Copilot and Claude automate worksheets, Power Query, DAX measures, VBA macros, PivotTables, charts, conditional formatting, slicers, named ranges and data connections through natural language. The project also ships a feature-parity CLI (excelcli) for coding agents, sharing the same core commands as the MCP server. It requires Windows with Excel 2016 or later installed, and all Excel files must be closed before use since the server needs exclusive access to workbooks.
Tools
range
Reads and writes cell values, formulas, number display formats, and data validation for a range.
range_format
Applies visual styling — fonts, fills, borders — and column auto-fit to a range.
Setup
1) VS Code users: install the extension from the marketplace (one-click, recommended). 2) Claude Desktop users: download the .mcpb file from the latest GitHub release and install it. 3) Any other MCP client: download mcp-excel.exe from the latest release and add it to PATH. 4) Alternatively install via dotnet tool install --global Sbroenne.ExcelMcp.McpServer (requires the .NET 10 runtime), then auto-configure your agent with npx add-mcp "mcp-excel" --name excel-mcp. 5) Before use, close all open Excel files and make sure Excel 2016 or later is installed locally.
Fit and risk
Not for
- Server-side data processing (use libraries like ClosedXML or EPPlus instead)
- Linux or macOS users (Windows plus a local Excel install is required)
- High-volume batch processing workloads
Required permissions
- Controls the local Excel.Application process via Windows COM automation
- Requires exclusive access to the Excel workbook files it operates on
- Creates and modifies workbook files (e.g. .xlsx) on disk during automation
Risks and side effects
- Drives a real local Excel process and edits real workbooks, so files should be backed up or version-controlled before automation
- Supports running VBA macros, which can execute arbitrary code if the macro source is untrusted
- Not an official Microsoft product — maintained by an independent community developer, not Excel or MCP's upstream owner
- Requires exclusive workbook access, which can conflict with other processes that have the file open
Troubleshooting
- Confirm you're on Windows with Excel 2016 or later installed
- Close all open Excel workbook files before running the server to avoid exclusive-access conflicts
- If installed via dotnet tool, verify the .NET 10 runtime is present
- If the npx add-mcp auto-configuration step fails, verify Node.js is installed
- For high token usage, consider switching to the CLI (excelcli) instead of the full MCP tool set
Use cases
Creating new Excel workbooks and populating them with sample data or tables
Importing data with Power Query, loading it to the Data Model, and creating DAX measures
Building PivotTables, charts, and interactive slicers from existing data
Formatting ranges (number formats, conditional formatting, styling) and converting them to Excel Tables
Running or managing VBA macros, and exporting Power Query M code for version control
Supported clients
Claude DesktopFull support
VS CodeFull support
GitHub CopilotFull support