SQLBot
A conversational data analysis system that turns natural-language questions into SQL using a large language model and retrieval-augmented generation.
SQLBot lets a user ask questions in natural language and returns data and visualization charts by generating SQL against a connected data source, using RAG to improve accuracy, and supports further guided analysis on the results. Configuration only requires pointing it at a model and a data source, without additional development.
It provides workspace-level resource isolation and fine-grained data permissions, and can be embedded into other applications through a web widget, a popup, or an MCP connection, including integration with tools such as n8n, Dify, and MaxKB. Custom prompts and a terminology library let a team calibrate SQL generation to its own domain, and accuracy is intended to improve as usage data accumulates.
It deploys as a Docker container with a bundled PostgreSQL database and is developed by the team behind the DataEase open-source BI tool.