SQL Copilot

SQL Copilot is an agentic AI to SQL interface powered by LangGraph, featuring knowledge base management and self-evolution capabilities. The system not only allows users to query databases using natural language but also continuously learns from historical interactions to optimize knowledge base structure and query performance.

video.mp4

✨ Core Features

🧠 Knowledge Base Management

• Intelligent Knowledge Storage: Automatically stores successful query patterns, table sql
• Hybrid Search RAG: Conducts multi vector similarity search with a rerank for rearrangement
• Pattern Optimization: Dynamically adjusts knowledge base weights based on query frequency and success rates

💡 Intelligent Query

• Natural Language Understanding: Converts complex user questions into accurate SQL queries
• Graph Workflow: Uses state graphs to manage query analysis, intent recognition, SQL generation, and error recovery
• Schema Awareness: Automatically understands database structure and relationships
• Human-in-the-Loop: Actively incorporates human feedback and guidance throughout the query process
• Self-Correction: Automatically fixes invalid SQL queries and learns correction strategies

⚡ Technical Dependencies

• LangGraph: Workflow orchestration
• LangChain: LLM integration
• Milvus: Vector database (for knowledge base storage)
• SQLAlchemy: Database abstraction
• Pandas: Data manipulation

⚡ Dependencies

• LangGraph: Workflow orchestration
• LangChain: LLM integration
• Milvus: Vector database
• SQLAlchemy: Database abstraction
• Pandas: Data manipulation

🚀Quick Start

1. Environment Configuration

Configure LLM Service

Create configuration file config/llm_config.json:

{
  "chat": {
    "model_name": "gpt-4",
    "api_key": "your-openai-api-key",
    "base_url": "https://api.openai.com/v1",
    "temperature": 0.1,
    "max_tokens": 2000
  }
}

Configure Database Connection

Create config/database_config.json:

{
  "chenjie": {
    "name": "Example Database",
    "type": "mysql",
    "host": "localhost",
    "port": 3306,
    "username": "your_username",
    "password": "your_password",
    "database": "your_database"
  }
}

5. Initialize System

# Initialize database structure
python -c "from app import init; init('chenjie')"

6. Launch Web Application

python web_app.py

# Or use interactive mode
python app.py

7. Start Using

Open your browser and navigate to http://localhost:5123. Enter your natural language queries in the chat interface, for example:

• "Query the product with the highest sales"
• "Analyze sales trends across regions"
• "Find the most active users from last month"