MCP Data Analyst

A lightweight Model Context Protocol (MCP) server built with Python and Pandas that exposes reusable data-analysis capabilities as MCP tools.

Overview

This project demonstrates how an MCP-compatible client can discover and invoke external data-analysis tools through a standardized MCP server.

The server analyzes a sample sales CSV dataset and exposes five tools:

• dataset_info - Returns dataset dimensions and column names

• dataset_summary - Returns descriptive statistics

• missing_values - Detects missing values

• sales_by_region - Filters sales records by region

• column_analysis - Calculates statistics for numeric columns

Related MCP server: Claude Data Buddy

Architecture

User / AI Host | v MCP Client | v MCP Server | v MCP Tools | v Python + Pandas | v Sales CSV Dataset

Project Structure

mcp-data-analyst/ ├── data/ │ └── sample_sales.csv ├── tests/ │ └── test_analysis.py ├── tools/ │ ├── init.py │ ├── analysis_tools.py │ └── csv_tools.py ├── server.py ├── requirements.txt ├── pytest.ini ├── .gitignore └── README.md

Technologies

• Python

• Model Context Protocol (MCP)

• Pandas

• Pytest

• MCP Inspector

Installation

Create and activate a virtual environment.

Windows PowerShell:

python -m venv venv
.\venv\Scripts\Activate.ps1