home / mcp / excel analytics mcp server

Excel Analytics MCP Server

Locally upload Excel/CSV data, query with natural language, and run reusable analyses via a local dashboard.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "blakethom8-excel-mcp": {
      "command": "uv",
      "args": [
        "run",
        "python",
        "-m",
        "excel_mcp"
      ],
      "env": {
        "EXCEL_MCP_HOME": "YOUR_PATH_HERE"
      }
    }
  }
}

You deploy this MCP server locally to upload Excel or CSV data, question it in natural language, and save reusable analyses without leaving your machine. It creates a personal toolbox of data queries and tools that you can run again and again, all through a local REST dashboard and a tightly integrated runtime.

How to use

You operate the MCP server from your computer and connect to it with an MCP client. Start the server, open the local dashboard in your browser, upload data, run analyses, and save useful queries as reusable tools. You can create custom tools that Claude can use on your behalf and manage all tools from the dashboard.

How to install

Prerequisites: Python 3.10 or newer, a supported operating system (macOS, Windows, Linux), and a working Claude Desktop installation.

# Option 1: Install via uvx (recommended)
# Install uv if you don't have it
curl -LsSf https://astral.sh/uv/install.sh | sh

# Install and configure Claude Desktop
uvx excel-analytics-mcp --setup

Restart Claude Desktop after the setup and then start using the server from the dashboard.

Option 2: Install from source

git clone https://github.com/blakethom8/excel-mcp.git
cd excel-mcp
bash install.sh

The install script will perform these steps: - Check for Python 3.10+ and install uv if needed - Install all dependencies in an isolated virtual environment - Configure Claude Desktop automatically - Create the data directory at ~/Documents/Excel Analytics/

Additional setup notes

You can override the base data directory with the EXCEL_MCP_HOME environment variable. The default data layout places data, tools, and configurations under ~/Documents/Excel Analytics/.

Security and data handling

All data remains on your machine. The MCP server stores data locally in the configured directory. Read-only SQL queries are supported to protect data integrity. Python tools run in a sandboxed environment with restricted imports.

Configuration

The configuration file resides at ~/Documents/Excel Analytics/config.json and can be edited to adjust runtime behavior. The dashboard runs on port 8765 by default. You can adjust this in the config if needed.

Troubleshooting

If the dashboard does not appear or data does not load, verify: - The MCP server is running and connected to Claude Desktop - The REST API is listening on localhost:8765 - Your data directory contains the expected database and tool files

Notes

The MCP server supports a set of core tools for data exploration and analysis, and it can register dynamic tools stored as JSON files. Tools persist across sessions and can be shared across the same installation.

Available tools

list_datasets

Show all loaded tables with row and column counts.

describe_dataset

Provide column names, types, sample values, and basic statistics for a dataset.

query

Execute read-only SQL queries against the uploaded data.

summarize

Generate a statistical summary for a table or a specific column.

save_analysis

Save a SQL query as a reusable named tool for later use.

create_tool

Build a custom Python tool in a sandboxed environment.

list_my_tools

Display all saved tools you have created.

edit_tool

Update the parameters or code of an existing tool.

delete_tool

Remove a saved tool from your toolkit.

test_tool

Run a tool with test parameters to verify its behavior.