MCP Directory

How to add MCP Server for Data Exploration to Claude Desktop

Your personal data scientist assistant — explore CSV datasets with Claude through autonomous Python analysis. Paste the config into ~/Library/Application Support/Claude/claude_desktop_config.json and restart Claude Desktop.

Last updated June 14, 2026 · 543 · stdio · no auth

Claude Desktop config for MCP Server for Data Exploration

uvx mcp-server-ds
{
  "mcpServers": {
    "mcp-server-for-data-exploration": {
      "command": "uvx",
      "args": [
        "mcp-server-ds"
      ]
    }
  }
}

Requires `uv` (the Python package runner). Install it from https://docs.astral.sh/uv/ if `uvx` is not found.

Setup steps

  1. 1Open Claude Desktop → Settings → Developer → Edit Config (this opens ~/Library/Application Support/Claude/claude_desktop_config.json).
  2. 2Paste the MCP Server for Data Exploration config below under the top-level "mcpServers" key.
  3. 3Fill in any placeholder secrets (API keys, paths) in the snippet.
  4. 4Save the file, then fully quit and reopen Claude Desktop.
  5. 5Open a chat and confirm MCP Server for Data Exploration's tools appear under the 🔌 tools menu.

Before you start

  • Claude Desktop (or another MCP client)
  • uv / uvx (Python package runner) installed
  • Python environment for running analysis scripts

What MCP Server for Data Exploration can do in Claude Desktop

load-csv

Loads a CSV file into a DataFrame. Arguments: csv_path (string, required) — path to the CSV file; df_name (string, optional) — name for the DataFrame, defaults to df_1, df_2, etc.

run-script

Executes a Python script. Arguments: script (string, required) — the script to execute.

Security

The run-script tool executes arbitrary Python code on the host machine, and load-csv reads files from the local filesystem. Only run this server against data and scripts you trust, since the model can execute any Python it generates.

MCP Server for Data Exploration + Claude Desktop FAQ

Where is the Claude Desktop config file?

Claude Desktop reads MCP servers from ~/Library/Application Support/Claude/claude_desktop_config.json. Paste the MCP Server for Data Exploration config there under the "mcpServers" key and restart the client.

Is MCP Server for Data Exploration safe to use with Claude Desktop?

The run-script tool executes arbitrary Python code on the host machine, and load-csv reads files from the local filesystem. Only run this server against data and scripts you trust, since the model can execute any Python it generates.

How do I start exploring a dataset?

In Claude Desktop, select the explore-data prompt template from MCP, then provide the required inputs: csv_path (local path to the CSV file) and topic (the subject of exploration, e.g. "Weather patterns in New York").

How do I install it?

On macOS you can run `python setup.py` in the repo, or add the server to your Claude Desktop config. For the published package use command `uvx` with args `mcp-server-ds`; for local development use `uv` with `--directory` pointing at your checkout.

Is it safe to run?

The run-script tool executes arbitrary Python on your machine, so only use it with datasets and analyses you trust.

View repo Full MCP Server for Data Exploration page