Visidata Mcp

Visidata Mcp is an MCP server that provides access to VisiData tabular data operations along with automated statistical visualization and skills analysis tools. It connects Model Context Protocol clients like Claude Desktop and Cursor directly to VisiData's data processing engine, enabling data scientists, business analysts, and HR researchers to manipulate and inspect structured files without manual terminal interactions. Users can load, preview, filter, sort, and convert datasets across formats including CSV, JSON, Excel, SQLite, and Parquet. Beyond baseline spreadsheet manipulation, the server generates publication-ready statistical visualizations such as correlation heatmaps, distribution plots, histograms, and scatter plots using matplotlib and seaborn. It also includes dedicated utilities for labor market data analysis, allowing users to parse comma-separated skills strings into one-hot encodings, calculate regional skill frequencies, and evaluate salary distributions against multi-variable combinations. By bridging tabular file operations and graphical output generation, it simplifies exploratory data analysis directly from AI assistant interfaces.

Category: Data & Analytics

Tags: csv, data-exploration, tabular-data, visidata

Visit Visidata Mcp

How to install and configure Visidata Mcp

1. Install the server globally via npm by running npm install -g @moeloubani/visidata-mcp@beta (requires Python 3.10+), or install with Python using pip install visidata-mcp. 2. To configure Claude Desktop, edit ~/Library/Application Support/Claude/claude_desktop_config.json and add: json { "mcpServers": { "visidata": { "command": "visidata-mcp" } } } 3. To configure Cursor AI, create .cursor/mcp.json in your project folder with the same mcpServers entry pointing to visidata-mcp. 4. Completely restart your AI client application to load the tools.

What you can do with Visidata Mcp

• Convert complex tabular datasets across formats including CSV, JSON, Excel spreadsheets, SQLite tables, and Parquet files. - Generate publication-ready statistical visualizations, including correlation heatmaps, KDE plots, and categorized scatter graphs saved directly to disk. - Parse comma-separated skills strings from job postings into one-hot encoded columns for structured workforce analytics. - Analyze salary distributions across geographic locations and specialized skill sets to benchmark compensation trends. - Inspect, filter, and summarize large datasets using row sampling, conditional filtering, and column-level statistical profiling.

Key facts

• https://github.com/moeloubani/visidata-mcp
• Data & Analytics
• csv, data-exploration, tabular-data, visidata

Part of MCP Servers

How do I install Visidata Mcp?

You can install it globally via npm using the command npm install -g @moeloubani/visidata-mcp@beta, provided you have Python 3.10 or higher installed. Alternatively, you can install the package directly through Python by executing pip install visidata-mcp or cloning the repository and running pip install -e .

What can Visidata Mcp do?

Visidata Mcp connects AI assistants to VisiData tabular functionality. It loads and samples data, converts between formats such as CSV, JSON, and Excel, filters and sorts records, computes column statistics, renders statistical distribution and correlation plots, and performs job market skills and salary analysis.

Which MCP clients work with Visidata Mcp?

The server works with any Model Context Protocol compliant application. The official documentation provides explicit configuration examples for Claude Desktop via claude_desktop_config.json and Cursor AI via the project-level .cursor/mcp.json file using the visidata-mcp command.

What file formats does Visidata Mcp support?

Visidata Mcp supports a wide variety of formats including CSV, TSV, and Excel files; structured formats like JSON, JSONL, XML, and YAML; SQLite databases; scientific formats such as HDF5, Parquet, and Arrow; archives including ZIP and TAR; and HTML tables.

Is Visidata Mcp open source?

Yes, Visidata Mcp is open-source software released under the MIT License. The complete source code and licensing details are publicly accessible on GitHub at https://github.com/moeloubani/visidata-mcp.