aurelio-nakamura/dataloupe

Overview

dataloupe converts CSV, Parquet, or Excel files into a single self-contained interactive HTML file for offline exploration. It also acts as an MCP server that lets AI assistants query local data without sending it anywhere. Built in TypeScript.

Best for

Best for
Developers who need private, local data exploration and AI querying without uploading files.

Use cases

• Generate an interactive HTML explorer from a CSV file
• Let an AI assistant query local Parquet or Excel data via MCP
• Explore data offline without uploading files to a service

How to use

Install

npx dataloupe data.csv --open

Tools exposed

• list_data_files
• describe_data
• preview_data
• query_data
• sql_query
• visualize_data
• diff_data

Tested with

Claude Desktop, Cursor, VS Code

Notes

dataloupe converts CSV, Parquet, or Excel files into a single self-contained interactive HTML file for offline exploration. It also acts as an MCP server that lets AI assistants query local data without sending it anywhere. Built in TypeScript.

0 stars on GitHub. Last updated 2026-09-01. Licensed MIT.

Use cases

• Generate an interactive HTML explorer from a CSV file
• Let an AI assistant query local Parquet or Excel data via MCP
• Explore data offline without uploading files to a service

Pros

• Everything runs locally, so data never leaves your machine
• Output is a single HTML file that is easy to open or share
• Supports common formats like CSV, Parquet, and Excel

Cons

• Server mode requires an MCP-compatible AI client
• Interactive HTML capabilities beyond basic viewing are unspecified
• Currently has no GitHub stars or community traction

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.