A Model Context Protocol (MCP) server that lets you manipulate Excel files without needing Microsoft Excel installed. Create, read, write, and analyze Excel workbooks (.xlsx, .xlsm) with AI assistants like Claude. Complete Excel automation through LLM integration.
Key Features:
- ✅ Read and write Excel files without Microsoft Excel
- ✅ Full support for XLSX, XLSM, XLTX, XLTM formats
- ✅ Create charts, pivot tables, and dashboards
- ✅ Import/export CSV, JSON, SQL, PDF
- ✅ Works with Claude AI and other LLM assistants
- ✅ Cross-platform: Windows, macOS, Linux
- ✅ Easy installation via npm/npx
- 📊 Excel file operations: Read and write XLSX, XLSM, XLTX, XLTM files
- 📖 Data extraction: Read data from Excel sheets with pagination support
- ✍️ Write operations: Write data and formulas to Excel workbooks
- 📋 Sheet management: Create, delete, rename, and copy worksheets
- 📈 Charts and visualizations: Create charts, pivot tables, and dashboards
- 🔄 Data import/export: Import from CSV, JSON, SQL and export to multiple formats
- 🎨 Professional formatting: Automatic styling and formatting for Excel documents
- Read and write Excel files with full formatting support
- Create professional tables with automatic styling
- Generate charts and visualizations
- Import from CSV, JSON, and SQL sources
- Export to multiple formats (CSV, JSON, PDF)
- Automatic column width adjustment
- Rich text formatting and styling
- Professional color schemes and themes
- Publication-ready document generation
- Dynamic dashboards with multiple visualizations
- Template-based report generation
- Data filtering and analysis
- Pivot tables and advanced calculations
- Batch processing and automation
Security Score: 100/100 | Risk Level: Low
This project has been independently audited by MseeP.ai, providing ongoing security validation and trust assessment for the MCP ecosystem.
The easiest way to use Excel MCP Server is with npx (no installation required):
npx @guillehr2/excel-mcp-server@latestOr install globally:
npm install -g @guillehr2/excel-mcp-serverAdd to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"excel-master": {
"command": "npx",
"args": [
"-y",
"@guillehr2/excel-mcp-server@latest"
]
}
}
}{
"mcpServers": {
"excel-master": {
"command": "npx",
"args": [
"-y",
"@guillehr2/excel-mcp-server@1.0.7"
]
}
}
}{
"mcpServers": {
"excel-master": {
"command": "excel-mcp-server"
}
}
}If you're developing or want to run from source:
{
"mcpServers": {
"excel-master": {
"command": "node",
"args": ["path/to/Excel-MCP-Server-Master/index.js"]
}
}
}- Create Excel workbooks -
create_workbook_tool - Open Excel files -
open_workbook_tool(XLSX, XLSM, XLTX, XLTM) - Save Excel files -
save_workbook_tool - List Excel sheets -
list_sheets_tool - Manage worksheets -
add_sheet_tool,delete_sheet_tool,rename_sheet_tool
- Write to Excel -
write_sheet_data_tool,update_cell_tool - Read from Excel - Built-in data extraction with pagination
- Excel tables -
add_table_tool,create_formatted_table_tool - Excel formulas - Full formula support in all write operations
- Create Excel charts -
add_chart_tool(column, bar, line, pie, scatter) - Excel dashboards -
create_dashboard_tool - Pivot tables - Advanced data analysis
- Data filtering -
filter_data_tool
- Import to Excel -
import_data_tool(CSV, JSON, SQL) - Export from Excel -
export_data_tool(CSV, JSON, PDF) - PDF export -
export_single_sheet_pdf_tool,export_sheets_pdf_tool
# Create a new workbook with formatted data
result = create_formatted_table_tool(
file_path="sales_report.xlsx",
sheet_name="Q4 Sales",
start_cell="A1",
data=[
["Region", "Q4 Sales", "Growth %"],
["North", 125000, 15.2],
["South", 98000, 8.7],
["East", 156000, 22.1],
["West", 89000, -3.2]
],
table_name="Q4SalesData",
table_style="TableStyleMedium9",
formats={
"B2:B5": "#,##0", # Number format for sales
"C2:C5": "0.0%", # Percentage format
"A1:C1": {"bold": True, "fill_color": "366092"} # Header styling
}
)
# Add a chart based on the table data
chart_result = add_chart_tool(
file_path="sales_report.xlsx",
sheet_name="Q4 Sales",
chart_type="column",
data_range="A1:B5",
title="Q4 Sales by Region",
position="E2",
style="colorful-1"
)# Create a comprehensive dashboard
dashboard_result = create_dashboard_tool(
file_path="executive_dashboard.xlsx",
data={
"Data": [
["Month", "Revenue", "Expenses", "Profit"],
["Jan", 50000, 30000, 20000],
["Feb", 55000, 32000, 23000],
["Mar", 48000, 29000, 19000]
]
},
dashboard_config={
"tables": [
{
"sheet": "Dashboard",
"name": "MonthlyData",
"range": "Data!A1:D4",
"style": "TableStyleMedium9"
}
],
"charts": [
{
"sheet": "Dashboard",
"type": "line",
"data_range": "Data!A1:B4",
"title": "Revenue Trend",
"position": "E1",
"style": "dark-blue"
},
{
"sheet": "Dashboard",
"type": "column",
"data_range": "Data!A1:D4",
"title": "Monthly Comparison",
"position": "E15",
"style": "colorful-2"
}
]
}
)# Import data from multiple sources
import_result = import_data_tool(
excel_file="analysis.xlsx",
import_config={
"csv": [
{
"file_path": "sales_data.csv",
"sheet_name": "Sales",
"delimiter": ",",
"encoding": "utf-8"
}
],
"json": [
{
"file_path": "customer_data.json",
"sheet_name": "Customers",
"format": "records"
}
]
},
create_tables=True
)
# Filter and analyze the imported data
filtered_data = filter_data_tool(
file_path="analysis.xlsx",
sheet_name="Sales",
table_name="Table_Sales_1",
filters={
"Region": ["North", "South"],
"Sales": {"gt": 10000}
}
)The server automatically applies professional formatting:
- Column width adjustment based on content length
- Row height optimization for wrapped text
- Professional color schemes for charts and tables
- Consistent styling throughout documents
Extensive chart customization options:
- 50+ predefined styles (light, dark, colorful themes)
- Custom color palettes for brand consistency
- Professional layouts with proper spacing
- Multiple chart types: column, bar, line, pie, scatter, area
Create reports from templates:
- Reusable templates for consistent reporting
- Dynamic data substitution
- Automatic chart updates
- Format preservation
- Node.js 14.0 or higher
- Python 3.8 or higher
- Operating System: Windows, macOS, or Linux
Python dependencies are automatically installed on first run:
- fastmcp
- openpyxl
- pandas
- numpy
- matplotlib
- xlsxwriter
- xlrd
- xlwt
For detailed documentation, see:
We welcome contributions! Please see CONTRIBUTING.md for guidelines.
# Clone the repository
git clone https://github.com/guillehr2/Excel-MCP-Server-Master.git
cd Excel-MCP-Server-Master
# Install dependencies
npm install
pip install -r requirements.txt
# Run in development mode
node index.js- Python not found: Ensure Python 3.8+ is installed and in your PATH
- Dependencies fail to install: Try running with administrator privileges
- MCP client doesn't recognize the server: Restart your MCP client after configuration
For more help, see our troubleshooting guide or open an issue.
This project is licensed under the MIT License - see the LICENSE file for details.
- Built with FastMCP
- Excel manipulation powered by openpyxl
- Data processing with pandas
- Published on npm for easy distribution
- Special thanks to lwsinclair for the independent security audit and MseeP.ai integration
Made with ❤️ for the MCP ecosystem
If you find this Excel MCP Server useful, please consider giving it a ⭐ on GitHub!
Excel MCP Server by Guillem Hermida | GitHub | NPM | Contact: qtmsuite@gmail.com
Excel manipulation without Microsoft Excel - Model Context Protocol server for Claude AI and LLM integration