NL-SQL

Slack bot that lets you query a SQLite database using plain English. You type a question with /ask-data, it converts it to SQL using Google Gemini, runs it, and posts the results back in Slack.

How it works

1. User sends a slash command like /ask-data total revenue by region last week
2. LangChain + Gemini turns that into a SQL query against the sales_daily table
3. Query runs on a local SQLite database
4. Results come back formatted in Slack

Setup

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Create a .env file:

SLACK_BOT_TOKEN=xoxb-...
SLACK_APP_TOKEN=xapp-...
GEMINI_API_KEY=your-key-here

Seed the database:

python seed_db.py

Run:

python app.py

Project structure

• app.py — Slack bot entry point, handles the /ask-data command
• nl_to_sql.py — Converts natural language to SQL using Gemini
• db.py — Runs queries against SQLite
• formatter.py — Formats query results into Slack blocks
• seed_db.py — Creates and populates the sample database

Database schema

Single table sales_daily with columns: date, region, category, revenue, orders.

Requirements

• Python 3.10+
• A Slack app with socket mode enabled and the /ask-data slash command configured
• Google Gemini API key