🔍 Streamlit SQL Database Agent

A Streamlit application that leverages LangChain and various LLM providers (OpenAI, Groq, Google, Anthropic) to create a safe, production-grade SQL querying assistant. Easily connect to your SQL Server database, configure an LLM, and get natural language insights with query execution and analysis.

---

📋 Table of Contents

• Features
• Prerequisites
• Installation
• Configuration
  • Database Settings
  • LLM Settings
• Running the App
• Usage Guide
  • 1. Configure Database
  • 2. Initialize LLM
  • 3. Create SQL Agent
  • 4. Execute Queries
  • 5. View Database Info & History
• Project Structure
• Dependencies
• Contributing
• License

---

🚀 Features

• One-click connection testing to SQL Server via ODBC
• Support for multiple LLM providers: OpenAI, Groq, Google, Anthropic
• Safe SQL agent with enforced read-only rules
• Interactive, step-by-step query execution breakdown
• Live metadata and schema browsing
• Query history tracking and replay

---

📜 Prerequisites

• Python 3.8 or higher
• A SQL Server instance accessible over network
• API key for your chosen LLM provider (OpenAI, Groq, Google, or Anthropic)

---

⚙️ Installation

1. Clone the repository

   git clone https://github.com/ashifhusainoo7/SQLToNLP.git
   cd SQLToNLP

2. Create and activate a virtual environment (recommended)

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

3. Install dependencies

   pip install -r requirements.txt

---

🛠️ Configuration

All configuration is done through the Streamlit sidebar.

Database Settings

1. Server: Your SQL Server address (e.g., myserver\SQLEXPRESS)
2. Database: Target database name
3. Username & Password: Credentials with read-only permissions
4. Driver: Choose ODBC Driver 17 or 18 for SQL Server (or FreeTDS)

Click Test Database Connection to verify and initialize.

LLM Settings

1. LLM Provider: Select from OpenAI, Groq, Google, Anthropic
2. Model: Choose model compatible with provider (e.g., gpt-4-turbo)
3. API Key: Enter your provider API key
4. Temperature & Max Tokens (optional)

Click Initialize LLM to connect.

---

▶️ Running the App

Start the Streamlit server:

streamlit run streamlit_with_sql.py

This will open the UI in your default browser at http://localhost:8501.

---

🧭 Usage Guide

Once the app is running, follow these steps:

1. Configure Database

• Enter SQL Server details in the sidebar
• Click 🔗 Test Database Connection
• Upon success, the Database status indicator turns ✅

2. Initialize LLM

• Select LLM provider and model
• Enter your API key and adjust parameters
• Click 🚀 Initialize LLM
• LLM status indicator turns ✅

3. Create SQL Agent

• After database and LLM are ready, click ⚡ Create SQL Agent
• Agent status indicator turns ✅

4. Execute Queries

• In the Query Interface, type natural language questions (e.g., "List users who haven't logged in today with their company names")
• Click 🔍 Execute Query to send to the agent
• View results and the generated SQL in the "View Execution Steps" expander

5. View Database Info & History

• Refresh Table Info to load current schema: total tables, sample schemas
• Expand Available Tables to browse object names
• Query history appears in the sidebar with timestamps and statuses

---

🗂️ Project Structure

SQLToNLP/
├── streamlit_with_sql.py      # Main application file
├── requirements.txt          # Python dependencies
└── README.md                 # This documentation

---

📦 Dependencies

See requirements.txt for the full list:

streamlit
pandas
sqlalchemy
langchain
langchain-openai
langchain-anthropic
langchain-google-genai
langchain-groq
langgraph
langchain_community
pyodbc

---

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/YourFeature)
3. Commit your changes (git commit -m 'Add feature')
4. Push to branch (git push origin feature/YourFeature)
5. Open a Pull Request

---

📄 License

This project is licensed under the MIT License. See LICENSE for details.