SQL Agent (Node.js/TypeScript)

A powerful, multi-database and multi-LLM SQL assistant built with Node.js, Express, and React. It uses LangGraph.js to orchestrate intelligent agents that reason about your database schema, generate optimized SQL queries, and provide analytical summaries—all through a sleek, modern UI.

---

🌟 Features

Multi-Database Support

Powered by Knex.js, the agent seamlessly supports multiple database engines:

• PostgreSQL - Full support with SSL/TLS for secure remote connections
• MySQL - Complete MySQL 8+ compatibility
• SQLite - Local file-based databases for development and testing

Multi-LLM Provider Integration

Native integration with leading AI providers:

• OpenAI (Compatible) - Works with OpenAI and any OpenAI-compatible API (Groq, OpenRouter, Ollama, Mistral, etc.)
• Anthropic Claude - Direct integration with Claude models
• Google Gemini - Native Google Generative AI support

Modern UI Experience

• Dark Theme with Glassmorphism - Vibrant, professional design
• Real-time Reasoning Visualization - Watch the agent think step-by-step
• SQL Syntax Highlighting - Prism-based code highlighting
• Query Status Tracking - Visual indicators for pending, approved, denied, and executed queries
• Responsive Design - Works seamlessly on desktop and mobile
• Configuration Persistence - Settings saved to localStorage

---

🛠️ Installation

Prerequisites

• Node.js v18+ (v20 recommended)
• npm v9+ or yarn
• A database (PostgreSQL, MySQL, or SQLite)
• An LLM API key (OpenAI, Anthropic, Google, or compatible provider)

1. Clone the Repository

git clone https://github.com/varunreddy/Node-SQL-agent.git
cd Node-SQL-agent

2. Install Dependencies

Install dependencies for both the backend and frontend:

# Install root (server) dependencies
npm install

# Install client dependencies
cd client && npm install && cd ..

3. Environment Setup

Create a .env file in the root directory:

touch .env

Add your API keys and optional configurations:

# ===== LLM Provider Keys =====
# At least one is required
OPENAI_API_KEY=sk-your-openai-key
ANTHROPIC_API_KEY=sk-ant-your-anthropic-key
GOOGLE_API_KEY=your-google-api-key
GROQ_API_KEY=your-groq-api-key

# ===== Optional: Custom OpenAI-compatible endpoint =====
OPENAI_BASE_URL=https://api.openai.com/v1

# ===== Optional: Default Model Settings =====
MODEL_NAME=gpt-4o
TEMPERATURE=0

# ===== Optional: Database Defaults =====
# These can also be configured in the UI
DATABASE_URL=postgresql://user:password@localhost:5432/mydb
SQLITE_PATH=./database.sqlite

---

🖥️ Usage

Development Mode

Start both the backend server and frontend dev server concurrently:

npm run dev

This runs:

• Backend: http://localhost:3001 (Express API server)
• Frontend: http://localhost:5173 (Vite dev server with HMR)

Production Build

# Build both server and client
npm run build

# Start production server
npm start
# or with environment flag
NODE_ENV=production node dist/server.js

The production server serves both the API and the static frontend from client/dist/.

---

⚙️ Configuration

Database Configuration (via UI)

1. Open the sidebar by clicking the toggle button
2. Select the Database tab
3. Choose your database engine:

PostgreSQL / MySQL

Field	Description
Host	Database server hostname (e.g., localhost, db.example.com)
Port	Connection port (PostgreSQL: 5432, MySQL: 3306)
Database	Database name
Username	Database user
Password	User password
Enable SSL/TLS	Toggle for secure remote connections (required for most cloud databases)

SQLite

Field	Description
DB Path	Local file path (e.g., ./database.sqlite, /data/mydb.db)

💡 Tip: Click Save after configuring to persist settings.

LLM Configuration (via UI)

1. Select the LLM Setup tab in the sidebar
2. Configure your AI provider:

Field	Description
Provider	Select: OpenAI (Compatible), Anthropic, or Google Gemini
Base URL	API endpoint (for OpenAI-compatible only). Use shortcuts for Groq, OpenRouter, Moonshot, Ollama
API Key	Your provider's API key
Model Name	Model identifier (e.g., gpt-4o, claude-3-5-sonnet-20240620, gemini-1.5-pro)
Max Tokens	Maximum response length
Temperature	Creativity level (0 = focused, 1 = creative)

Provider-Specific Notes

Provider	Base URL	Example Models
OpenAI	https://api.openai.com/v1 (default)	gpt-4o, gpt-4o-mini, gpt-4-turbo
Groq	https://api.groq.com/openai/v1	llama-3.3-70b-versatile, mixtral-8x7b-32768
OpenRouter	https://openrouter.ai/api/v1	openai/gpt-4o, anthropic/claude-3.5-sonnet
Ollama	http://localhost:11434/v1	llama3, mistral, codellama
Anthropic	Native (no URL needed)	claude-3-5-sonnet-20240620, claude-3-opus-20240229
Google Gemini	Native (no URL needed)	gemini-1.5-pro, gemini-1.5-flash

---

🔐 SSL/TLS for Remote Databases

For secure connections to cloud databases (AWS RDS, Azure Database, Neon, Supabase, etc.):

1. Enable SSL Toggle in the Database configuration panel
2. The client uses rejectUnauthorized: false by default for compatibility with self-signed certificates

For production with certificate validation:

# Option 1: Use sslmode in connection string
DATABASE_URL=postgresql://user:pass@host:5432/db?sslmode=require

# Option 2: Provide CA certificate
SSL_CERT=/path/to/ca-certificate.crt
SSL_CLIENT_CERT=/path/to/client-cert.crt  # For mTLS
SSL_CLIENT_KEY=/path/to/client-key.key    # For mTLS

---

🚀 Deployment

Vercel (Recommended)

The project includes a vercel.json configuration for seamless deployment:

1. Push your code to GitHub
2. Import the repository in Vercel
3. Add environment variables in Vercel dashboard
4. Deploy!

Vercel will automatically:

• Run npm run build (compiles TypeScript + builds React)
• Start npm start (Node.js serverless function)
• Serve static files from client/dist/

Docker

# Dockerfile example
FROM node:20-alpine
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build
EXPOSE 3001
CMD ["npm", "start"]

---

🛡️ Security Considerations

1. API Keys: Never commit .env files. Use environment variables in production.
2. SQL Injection: The agent uses parameterized queries via Knex.js.
3. SSL/TLS: Always enable for remote database connections.
4. CORS: Configured for local development; restrict in production.
5. Rate Limiting: Consider adding rate limiting for production deployments.

---

🧪 Development

Project Scripts

Script	Description
npm run dev	Start dev servers (backend + frontend concurrently)
npm run dev:server	Start backend only (with tsx hot-reload)
npm run dev:client	Start frontend only (Vite dev server)
npm run build	Build for production
npm start	Run production server
npm run lint	Run ESLint (in client/)

Tech Stack

Backend:

• Express.js - Web server
• LangGraph.js - Agent orchestration
• LangChain.js - LLM integrations
• Knex.js - SQL query builder
• Zod - Schema validation
• TypeScript

Frontend:

• React 19 - UI framework
• Vite - Build tool
• Tailwind CSS 4 - Styling
• Lucide React - Icons
• React Syntax Highlighter - Code highlighting

---

🤝 Contributing

Contributions are welcome! Please:

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

---

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Built with ❤️ by Varun Reddy