Data Tools
The GenAI-powered toolkit for automated data intelligence.
Last verified:
What is Data Tools?
Intugle Data Tools is an open-source GenAI-powered Python library that helps data teams build a semantic layer over fragmented and siloed datasets. It auto-profiles and links isolated datasets, generates a business glossary from raw tables, creates smart SQL and reusable data products, and enables semantic search and natural language queries on top of your data. The tool uses DuckDB for in-memory data processing and Qdrant for semantic search on the semantic layer.
Key features include a Semantic Data Model that transforms fragmented datasets into an intelligent semantic graph, Business Glossary & Semantic Search capabilities, Conceptual Search (an AI-powered feature that generates data product plans from natural language queries), a built-in MCP (Model Context Protocol) server that exposes your data environment as tools for AI assistants, and a Streamlit-powered UI for visual, no-code semantic data modeling. The MCP server provides tools like get_tables and schema retrieval that allow AI assistants to discover and understand data structure.
This tool is perfect for Data Teams delivering Data Products to Business, Engineering Teams implementing Natural Language to SQL, data engineers, analysts, product managers, and even non-technical users who want to explore semantic data modeling without writing code. It's designed for organizations looking to turn messy, scattered data into a meaningful and connected knowledge layer so humans and AI can speak the same data language.
Data Tools pricing
Pricing model: Freemium
Intugle Data Tools is completely free and open-source. It is available on GitHub at github.com/Intugle/data-tools and on PyPI as pip install intugle. There is no paid tier for the open-source library - everyone can use it for free. The enterprise version at intugle.ai offers additional features like connecting 50+ data sources and enterprise AI data platform capabilities, but the open-source data-tools library itself has no cost.
Data Tools pros
- Open-source and free to use for everyone
- Auto-profiles and links siloed datasets automatically
- Generates business glossary from raw tables
- Creates smart SQL queries automatically
- Builds reusable data products
- Powers semantic search on data
- Enables natural language queries on data
- Uses DuckDB for fast in-memory processing
- Uses Qdrant for semantic search capabilities
- Includes built-in MCP server for AI assistant integration
- Streamlit UI allows no-code semantic modeling
- Conceptual Search generates data product plans from text
- Connects to 50+ data sources in enterprise version
- Exposes get_tables and schema tools for AI agents
- Portable semantic layer exportable as YAML files
- Great for Vibe Coding and conversational workflows
- Active open-source community on GitHub and Discord
Data Tools cons
- Conceptual Search is experimental and API may change
- Primarily Python-based requiring coding knowledge for full features
- Streamlit UI is newer and may have limited features vs code
- MCP server requires local setup on localhost:8080
- Semantic model must be generated before MCP tools work
- No official commercial support for open-source version
- Learning curve for semantic modeling concepts
- Enterprise features like 50+ source connections are separate
- Qdrant dependency adds infrastructure complexity
- New project with evolving documentation
Frequently asked questions about Data Tools
Is Intugle Data Tools free to use?
Yes! It is an open-source project available for everyone. You can install it via PyPI with pip install intugle and use all features without any cost.
What programming language is used for Intugle Data Tools?
Intugle Data Tools is a Python library. It is built as an open-source Python package that you can install and use in your Python projects.
What is the Semantic Data Model?
The Semantic Data Model transforms fragmented datasets into an intelligent semantic graph. It auto-profiles and links your siloed datasets, generates a business glossary from raw tables, and creates a connected knowledge layer that humans and AI can both understand.
What is Conceptual Search?
Conceptual Search is an AI-powered experimental feature that allows you to generate a data product plan from a natural language query. It bridges the gap between high-level business questions and concrete, executable data product definitions.
What is the MCP Server in Intugle Data Tools?
The built-in MCP (Model Context Protocol) server exposes your data environment as tools that AI assistants and LLM-powered clients can understand and use. It runs on localhost:8080 by default and provides tools like get_tables and schema retrieval for AI agents.
How do I install Intugle Data Tools?
You can install it via PyPI using the command 'pip install intugle'. The GitHub repository is at github.com/Intugle/data-tools where you can also find documentation and contribute to the project.
What is the Streamlit UI for?
The Streamlit-powered UI provides a visual, interactive web interface for building and visualizing semantic data models without touching any code. You can load data, explore relationships between tables, understand the semantic model, and download it as YAML files.
What databases does Intugle use internally?
Intugle Data Tools uses DuckDB for in-memory data processing and Qdrant for semantic search on the Semantic Layer. These power the fast processing and intelligent search capabilities.
Who is Intugle Data Tools perfect for?
It is perfect for Data Teams delivering Data Products to Business, Engineering Teams implementing Natural Language to SQL, data engineers, analysts, product managers, and anyone with no technical background who wants to explore semantic data modeling through the UI.
How do I connect AI assistants to my data with Intugle?
After building your SemanticModel, start the MCP server with 'intugle-mcp' command. Then connect from any MCP-compatible client like Cursor, VS Code, Gemini CLI, JetBrains AI Assistant, or Claude Code using the endpoint http://localhost:8080/semantic_layer/mcp.