home / mcp / data wrangling mcp server

Data Wrangling MCP Server

Provides data aggregation and descriptive statistics to preprocess and analyze datasets within MCP workflows.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "shibuiwilliam-mcp-server-data-wrangler": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/mcp-data-wrangler",
        "run",
        "mcp-data-wrangler"
      ]
    }
  }
}

You can run an MCP server dedicated to Data Wrangling tasks, enabling standardized data preprocessing, transformation, and analysis within the MCP ecosystem. This server focuses on aggregating data and computing descriptive statistics, making it easier to integrate wrangling steps into your workflows.

How to use

To use this MCP server, connect your MCP client to the provided MCP endpoint or start the local stdio server as described. You will be able to perform data wrangling operations such as aggregation and descriptive statistics on your datasets, then feed the results into downstream MCP pipelines.

How to install

Prerequisites you need before starting:

Install Node.js and npm on your machine to access the MCP inspector tooling.

Clone the project repository to your local machine and navigate into the project directory.

Additional setup steps

Option 1: Run as an MCP server using the local runtime (stdio) with uv. This starts the server in-process and exposes an MCP endpoint for you to connect a client to.

Option 2: Launch the MCP inspector to debug or explore the Data Wrangling server and then run it via the inspector if you prefer an interactive debugging flow.

Notes and configuration

The Data Wrangling MCP server provides two core capabilities: data aggregation and descriptive statistics. Ensure your client is configured to send appropriate data structures for wrangling and that the environment has access to the datasets you intend to process.

Available tools

aggregate_data

Performs data aggregation operations to summarize datasets across groups, time windows, or custom keys as part of the data wrangling process.

describe_statistics

Computes descriptive statistics such as mean, median, standard deviation, and percentiles to characterize data distributions.