Skip to main content

AI Data Analysis MCP

Project description

Zaturn: Your Co-Pilot For Data Analytics & Business Insights

Zaturn let's you analyze your data using AI chat; without needing you to write SQL/Python code or fiddling with aesthetically pleasing (but overwhelming) dashboards.

You can add Zaturn MCP to Claude Desktop (or any MCP client), connect your data sources, ask questions in natural language, and get instant insights with visualizations. With Zaturn, your AI can automatically understand the kind of data you have, query it, and give you useful pointers with a coherent narrative. You can ask specific questions like "Who is our most valuable customer?", or let AI explore your data with a question like "Here's all the data we have, give us some ideas for the next quarter."

PyPI Downloads

Join The Discord

Zaturn MCP server

But I can just upload my data to ChatGPT and ask it to analyze right?

Yes, but ChatGPT has an upload size limit of approximately 50MB for CSV files/spreadsheets, and uploading also takes time. Plus, it can't read data straight from your MySQL or PostgreSQL server. Zaturn can overcome all of these limitations, without moving your data anywhere. It simply equips your AI with SQL and visualization capabilities so AI can query your data directly, get the results, process them, and give you instant insights. With Zaturn, AI does not need to process your full dataset and keep it in its memory to answer your questions about the data.

Zaturn in Action

https://github.com/user-attachments/assets/d42dc433-e5ec-4b3e-bef0-5cfc097396ab

Features:

Multiple Data Sources

Zaturn can currently connect to the following data sources:

  • SQL Databases: PostgreSQL, SQLite, DuckDB, MySQL, ClickHouse
  • Files: CSV, Parquet

Connectors for more data sources are being added.

Visualizations

In addition to providing tabular and textual summaries, Zaturn can also generate the following image visualizations

  • Scatter and Line Plots
  • Histograms
  • Strip and Box Plots
  • Bar Plots

NOTE: The visuals will be shown only if your MCP client supports image rendering (e.g. Claude Desktop)

If you MCP client does not support images (e.g. Cursor) add the --noimg argument in the MCP config. Then the plots will be stored as files and the file location will be returned. You can view the plots with your file browser.

More visualization capabilities are being added.

Installation & Setup

  1. Install uv

  2. Install Zaturn with uv:

uv tool install zaturn
  1. Add to MCP config, with data sources:
"mcpServers": {
  "zaturn": {
    "command": "zaturn_mcp",
    "args": [
      "postgresql://username:password@host:port/dbname",
      "mysql+pymysql://username:password@host:3306/dbname",
      "sqlite:////full/path/to/sample_dbs/northwind.db",
      "/full/path/to/sample_dbs/titanic.parquet",
      "/full/path/to/sample_dbs/ny_aq.csv",
      "/full/path/to/sample_dbs/duckdb_sample.duckdb"
    ]
  },
}

OR add a sources.txt to the Zaturn config directory:

postgresql://username:password@host:port/dbname
mysql+pymysql://username:password@host:3306/dbname
clickhouse://username:password@host:port/dbname
sqlite:////full/path/to/sample_dbs/northwind.db
/full/path/to/sample_dbs/titanic.parquet
/full/path/to/sample_dbs/ny_aq.csv
/full/path/to/sample_dbs/duckdb_sample.duckdb

This file needs to be at ~/.config/zaturn/sources.txt on Linux/MacOS and at %APPDATA%\zaturn\sources.txt on Windows.

If your MCP client does not support image rendering, add the --noimg argument:

...
    "args": [
      "--noimg",
      "mysql+pymysql://username:password@host:3306/dbname",
...
  1. Set a system prompt if your LLM/IDE allows you to:
You are a helpful data analysis assistant. Use only the tool provided data sources to process user inputs. Do not use external sources or your own knowledge base.
  1. Ask a question and watch the magic:
User: List the top 5 customers by revenue for Northwind
AI: 
The top 5 customers by revenue for Northwind are:

1. B's Beverages with a revenue of $6,154,115.34
2. Hungry Coyote Import Store** with a revenue of $5,698,023.67
3. Rancho grande with a revenue of $5,559,110.08
4. Gourmet Lanchonetes with a revenue of $5,552,597.90
5. Ana Trujillo Emparedados y helados with a revenue of $5,534,356.6

Roadmap

  • Support for more data source types
  • More data visualizations
  • Predictive analysis and forecasting, e.g.:
Based on the revenue of the last 3 months, forecast next month's revenue.
  • Generate Presentations & PDFs
Manager: 
  I need a presentation to show the boss. Can you do it by EOD?
Analyst: 
  EOD?! Are you still in the 2010s? 
  I can get it done right now. Actually, you can do it right now.
  You know what? The boss can do it right now.
  • A native notebook interface

Help And Feedback

Raise an issue or join the Discord.

Support The Project

If you find Zaturn useful, please support this project by:

Your support will enable me to dedicate more of my time to Zaturn.

Example Dataset Credits

The pokemon dataset compiled by Sarah Taha and PokéAPI has been included under the CC BY-NC-SA 4.0 license for demonstration purposes.

Star History

Star History Chart

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

zaturn-0.1.6.tar.gz (150.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

zaturn-0.1.6-py3-none-any.whl (45.1 kB view details)

Uploaded Python 3

File details

Details for the file zaturn-0.1.6.tar.gz.

File metadata

  • Download URL: zaturn-0.1.6.tar.gz
  • Upload date:
  • Size: 150.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.6.12

File hashes

Hashes for zaturn-0.1.6.tar.gz
Algorithm Hash digest
SHA256 8107a0d638db743fb18494be037577b71abf673807b0e477f91afe592fd39111
MD5 98762b4602714cf3bfbf3e1686785c80
BLAKE2b-256 d06c634ae360a861a016bf97242e97a79424b51ad2f8910bd7595709be6160e3

See more details on using hashes here.

File details

Details for the file zaturn-0.1.6-py3-none-any.whl.

File metadata

  • Download URL: zaturn-0.1.6-py3-none-any.whl
  • Upload date:
  • Size: 45.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.6.12

File hashes

Hashes for zaturn-0.1.6-py3-none-any.whl
Algorithm Hash digest
SHA256 20ddde2310ecd672d9310aecda33d839c2b4ab84f8b6e995945330ba65fcd5a6
MD5 aecdd5e4a734918d338ae4f3c9f0af2b
BLAKE2b-256 539c64832b654fbf168b8fe0bbdfbdb23f45121ab526b3805064f738c76fcb63

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page