Claude Data Buddy

MCP server for analyzing CSV and Parquet files via natural language with Claude Desktop integration.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "iramk11-claude-data-buddy": {
      "command": "python",
      "args": [
        "/path/to/claude-data-buddy/main.py"
      ]
    }
  }
}

Claude Data Buddy is a Model Context Protocol (MCP) server that analyzes CSV and Parquet files through natural language interactions. It enables conversational data analysis and easy integration with Claude Desktop, letting you ask questions about your data and get structured results.

How to use

Use an MCP client to connect to Claude Data Buddy and perform data analyses with natural language prompts. You can ask the server to summarize files, describe data, run comprehensive analyses, and retrieve sample rows or statistics. Typical workflows include listing available data assets, requesting CSV or Parquet summaries, and asking for multi-step analyses that cover data types, null counts, and memory usage.

How to install

Prerequisites you need on your machine: Python 3.8 or newer. If you have a CUDA-compatible GPU, you can enable optional GPU-backed operations.

Step 1. Install Python dependencies.

pip install -r requirements.txt

Step 2. Run the MCP server directly.

python main.py

Step 3. Or run via the provided launcher for Claude Desktop integration.

./run_mcp_server.sh

Step 4. Optional Claude Desktop configuration. Use an MCP server entry that points to the Python runtime and the main server script.

{
  "mcpServers": {
    "claude_data_buddy": {
      "command": "python",
      "args": ["/path/to/claude-data-buddy/main.py"]
    }
  }
}

Additional sections

Configuration details are shown below for quick setup and reference. Ensure you use the exact command and arguments shown when wiring up your MCP client or Claude Desktop.

Project structure highlights include the MCP server implementation, client modes, and sample data files that you can use to verify functionality.

Available tools

list_data_files

Lists all available CSV and Parquet files in the data directory.

summarize_csv

Provides a comprehensive summary of a CSV file, including row/column counts, column names and data types, sample data, and basic statistics.

summarize_parquet

Provides a comprehensive summary for Parquet files, including structure and statistics.

analyze_csv

Performs multi-step analysis on a CSV file, including describe, head, columns, info, shape, and nulls.

comprehensive_analysis

Performs a complete multi-step analysis including summary statistics, data types, null value analysis, sample data, and memory usage.