DataFoundry Documentation

DataFoundry is an AI workbench for data analysis. It brings natural-language questions, data source management, read-only SQL execution, analysis traceability, and result delivery into a single workflow—so you can move through exploratory data analysis faster.

These docs are for product trials, customer demos, open-source visitors, and integration developers. Public documentation is available in English and Chinese.

Start here

Goal	Recommended reading
Understand product positioning and use cases	Product overview
Run a local demo end to end	Quick start
See what Web, TUI, and API support	Capabilities
Use the graphical workbench	Web workbench guide
Use the terminal interface	TUI guide
Connect your own data sources	Data sources guide
Browse supported data sources	Supported data sources
Learn about APIs and integration	REST API reference, Configuration API reference, and Agent Runtime reference
Understand system structure	Architecture overview
Review security boundaries	Security

Recommended path

For a first trial, read in this order:

1. Read Product overview to confirm the problem space and capability boundaries.
2. Follow Quick start to configure a model API key and run your first question against the built-in DTC Growth Review data source.
3. Read Capabilities to see coverage across Web, TUI, and backend API.
4. Choose Web workbench guide or TUI guide based on your entry point.
5. When you need your own data, read Data sources guide.

Documentation scope

This directory focuses on the public reading experience. It does not include project management notes, implementation plans, AI collaboration logs, historical refactor logs, or source-sensitive early discussions. Public docs describe only capabilities confirmed by current code, configuration, scripts, and local README files.

If you are extending or integrating the product, start with reference/ and architecture/. If you are just trying the product, overview.md, quick-start.md, capabilities.md, and guides/ cover the main paths.