explaingit

eosphoros-ai/db-gpt

Analysis updated 2026-06-21

18,736PythonAudience · dataComplexity · 3/5LicenseSetup · moderate

TLDR

DB-GPT is an open-source AI data assistant that lets you ask questions about your databases and spreadsheets in plain English, the AI writes and runs the SQL or code for you and returns charts or written reports.

Mindmap

mindmap
  root((db-gpt))
    What it does
      Natural language SQL
      Agentic analysis
      Data visualization
    Tech Stack
      Python
      Docker
    Data Sources
      Databases
      Spreadsheets
      CSV files
    Use Cases
      Business reporting
      Auto analysis
      Non-technical access
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Ask business questions about your database in plain English and have the AI write and execute the SQL automatically.

USE CASE 2

Connect CSV files or spreadsheets and let the AI generate charts and written summaries without writing any code.

USE CASE 3

Build reusable analysis workflows that non-technical teammates can run with one click for recurring reports.

USE CASE 4

Run multi-step data analysis where the AI plans, writes code, executes it in a sandbox, and compiles results into a report.

What is it built with?

PythonDocker

How does it compare?

eosphoros-ai/db-gptlss233/kirara-aispotify/luigi
Stars18,73618,75318,717
LanguagePythonPythonPython
Setup difficultymoderatehardeasy
Complexity3/54/53/5
Audiencedatadeveloperdata

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Requires an AI model backend configured, either a cloud API key (OpenAI, Anthropic, etc.) or a locally running model via Ollama.

Use freely for any purpose, including commercial, with no restrictions beyond keeping the copyright notice. MIT license.

In plain English

DB-GPT is an open-source AI assistant for working with data. Instead of writing database queries by hand, you can ask questions in plain English and the system figures out the SQL needed to answer them. It can connect to databases, spreadsheets, CSV files, and other data sources, then let an AI plan and execute multi-step analysis workflows automatically. The project is designed as an agentic system, meaning the AI can break a task into steps, write code or SQL at each step, run it in a sandboxed environment for safety, and compile the results into charts, dashboards, or written reports. Users can also create reusable "skills," which are packaged workflows for common analysis tasks that can be loaded and re-run whenever needed. DB-GPT supports multiple AI model providers and can be installed with a single command on macOS or Linux. Once running, it opens a web interface at a local address. The platform is aimed at individuals and teams who want to ask business questions of their data without needing a data analyst, from financial analysis to database profiling to CSV exploration. It is released under the MIT license and supports a wide range of AI backends.

Copy-paste prompts

Prompt 1
I installed DB-GPT and connected my PostgreSQL database. How do I ask it to show me total sales by region for last quarter as a bar chart?
Prompt 2
How do I create a reusable skill in DB-GPT for a monthly revenue report so non-technical teammates can run it without any SQL knowledge?
Prompt 3
I have a large CSV file of customer data. How do I load it into DB-GPT and ask it to find the top 10 customers by total spend?
Prompt 4
DB-GPT supports multiple AI backends. How do I configure it to use a local Ollama model instead of a cloud API key?
Prompt 5
Help me set up DB-GPT using Docker on macOS so I can start querying my data in plain English in under 10 minutes.

Frequently asked questions

What is db-gpt?

DB-GPT is an open-source AI data assistant that lets you ask questions about your databases and spreadsheets in plain English, the AI writes and runs the SQL or code for you and returns charts or written reports.

What language is db-gpt written in?

Mainly Python. The stack also includes Python, Docker.

What license does db-gpt use?

Use freely for any purpose, including commercial, with no restrictions beyond keeping the copyright notice. MIT license.

How hard is db-gpt to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is db-gpt for?

Mainly data.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Scan in gitsafehub Deploy in gitdeployhub eosphoros-ai on gitmyhub

Verify against the repo before relying on details.