DB-Communicator 💬

A LangChain-based conversational assistant for SQL databases and tabular data, powered by Google Gemini.

This app allows users to:

• Ask questions in plain English about SQL databases or uploaded CSV/XLSX files.
• Get SQL queries auto-generated, executed, and explained.
• Use retrieval-augmented generation (RAG) on tabular data for semantic answers.

---

🚀 Features

• 🧾 Natural language to SQL generation (via Gemini LLM)
• 🗃️ Query .db files or uploaded .csv/.xlsx
• 🔍 ChromaDB-powered RAG over tabular data
• 🧠 Gemini models (gemini-2.0-flash, gemini-embedding-exp-03-07) used for chat + embedding
• 🧑‍💻 Clean UI with Gradio, including file upload and multi-mode selection

---

🛠️ Setup

1. Clone this repo

git clone https://github.com/AviN27/DB-Communicator.git
cd DB-Communicator

2. Provide the required data

• XLSX or CSV files
• .sql or .db files

3. Run the Gradio chatbot

py src/app.py

📌 Notes

• Currently supports SQLite only.
• Optimized for small- to medium-sized tabular datasets.
• Gemini is used both for generating SQL and for semantic retrieval (RAG).

📣 Credits

• Developed with support from Youtube tutorials (Farzad from AI RoundTable)
• Flaticons