agno-agi

agno-agi / dash

Public

Self-learning data agent that grounds its answers in 6 layers of context. Inspired by OpenAI's in-house implementation.

1,731
193
100% credibility
Found Feb 02, 2026 at 363 stars 5x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Dash is a self-improving AI agent that answers natural language questions about database contents by generating grounded SQL queries and providing contextual insights.

How It Works

1
📖 Discover Dash

You hear about Dash, a friendly helper that chats with your data tables and gives smart answers to everyday questions.

2
🏠 Set it up locally

You follow easy steps to get Dash running on your own computer, like turning on a new app.

3
🏎️ Add sample racing data

You load fun Formula 1 racing stats so Dash has real info to play with right away.

4
💬 Connect to chat screen

You link Dash to a simple web chat page where you can type questions naturally.

5
❓ Ask your first question

You type something like 'Who won the most races?' and watch Dash think and respond.

6
🧠 Get insightful answers

Dash shares not just numbers, but helpful explanations and context about your data.

🌟 Dash learns and improves

With each question, Dash remembers tips and gotchas, getting smarter for your future asks.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 363 to 1,731 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is dash?

Dash is a Python self-learning data agent on GitHub that converts natural language questions into SQL queries against PostgreSQL databases, grounding answers in six context layers like schemas, business rules, validated queries, and runtime learnings. It delivers insights with interpretations, not just rows, and evolves by auto-saving fixes for errors like type quirks or date formats. Fire it up via Docker Compose, load data with CLI scripts, and query through FastAPI docs at localhost:8000 or CLI mode.

Why is it gaining traction?

Basic text-to-SQL agents falter on real schemas lacking meaning or tribal knowledge; Dash stands out with a self-learning loop that curates proven queries and discovers gotchas, cutting repeat failures without fine-tuning. Devs grab dash github examples for quick F1 demos, love injecting custom knowledge via JSON/SQL files, and appreciate the evals suite for accuracy checks. As a github dash python take on OpenAI's in-house self learning agent github, it hooks those building smarter data dashboards.

Who should use this?

Data analysts querying messy Postgres DBs for business insights, backend devs automating ad-hoc reports, or small teams replacing manual SQL with a self-improving github dash shell agent. Suited for revenue tracking, customer analytics, or any setup where schemas hide gotchas like misleading types.

Verdict

Try it for Postgres text-to-SQL with learning smarts—1560 stars and thorough docs with deploy scripts show promise, but 1.0% credibility score means it's early; validate via built-in evals before prod.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.