🤖 Multi-Agent AI System for Automated Database Insights

A production-ready multi-agent pipeline built with LangGraph and GPT-4o-mini that autonomously analyzes a SQLite database, answers data questions, and generates a downloadable PDF report — all orchestrated by a Supervisor agent.

---

🏗️ Architecture

User Prompt
     │
     ▼
┌─────────────┐
│  Supervisor  │  ← Routes between agents
└──────┬──────┘
       │
  ┌────┴────────────────────┐
  │          │              │
  ▼          ▼              ▼
Analyst   Expert        Reviewer
  │          │              │
Asks 10+  Queries DB    Summarizes +
questions  via SQL      generates PDF

Agent	Role	Tools
Supervisor	Orchestrates the pipeline	Structured routing
Analyst	Understands schema, asks insightful questions	get_schema
Expert	Answers questions by running SQL queries	get_schema, execute_sql
Reviewer	Summarizes findings, creates PDF report	generate_pdf_report

---

✨ Features

• 🔄 Fully autonomous multi-agent loop with LangGraph
• 🗄️ SQLite database with users & orders tables
• 📊 Analyst agent generates 10+ data-driven questions
• 🔍 Expert agent answers via live SQL queries
• 📝 Reviewer agent produces an 8-line summary + 2 actionable insights
• 📥 PDF report download via Streamlit UI
• 🔒 API key entered at runtime — never stored

---

🚀 Live Demo

Replace the link above after deploying to Streamlit Cloud.

---

🛠️ Local Setup

1. Clone the repo

git clone https://github.com/Ishaan20072612/Multi-Agent-AI-System-for-Automated-Database-Insights.git
cd Multi-Agent-AI-System-for-Automated-Database-Insights

2. Create a virtual environment

python -m venv venv
source venv/bin/activate      # Windows: venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Run the Streamlit app

streamlit run app.py

5. Open in browser

Go to http://localhost:8501 and enter your OpenAI API key in the sidebar.

---

☁️ Deploy to Streamlit Cloud (Free)

1. Fork or push this repo to your GitHub account
2. Go to share.streamlit.io and sign in with GitHub
3. Click New app → select this repo → set Main file path to app.py
4. Click Deploy — done in ~2 minutes!
5. Share the public URL with anyone

⚠️ Users will need to enter their own OpenAI API key in the sidebar. Your key is never embedded in the code.

---

📁 Project Structure

├── app.py              # Streamlit UI
├── agents.py           # Core LangGraph pipeline & agent logic
├── requirements.txt    # Python dependencies
└── README.md

---

🧪 Sample Database

users (15 rows): id, name, email, signup_date
orders (15 rows): id, user_id, amount, status, order_date

Sample questions the Analyst asks:

• Which users have placed the most orders?
• What is the average order value by month?
• What percentage of orders are pending vs completed vs cancelled?
• Which users signed up but never placed an order?

---

🔧 Tech Stack

Library	Version	Purpose
langchain	0.3.x	LLM framework
langchain-openai	0.3.x	OpenAI integration
langgraph	0.4.x	Agent orchestration
openai	1.x	GPT-4o-mini
fpdf	1.7.x	PDF generation
streamlit	1.x	Web UI

---

📄 License

MIT License — free to use, modify, and distribute.

---

👤 Author

Ishaan Chowdhury · @Ishaan20072612

Built With

• python