Oracle ADB AI Agent - Enterprise LangChain Integration

Enterprise-grade LangChain agent with Oracle Autonomous Database (ADB) integration using Model Context Protocol (MCP) for AI-powered database operations and API orchestration.

Overview

This system provides a sophisticated AI-powered interface for Oracle database operations, combining natural language processing with enterprise database connectivity. The agent can understand complex queries, execute SQL operations, explore schema relationships, and integrate with external APIs.

Key Features

AI-Powered Database Operations

• Natural language to SQL translation using OpenAI GPT-4o
• Intelligent query optimization and result interpretation
• Context-aware database exploration and schema discovery

Enterprise Oracle ADB Integration

• Production-ready Oracle Autonomous Database connectivity
• Wallet-based secure authentication
• Connection pooling for high-performance operations
• Transaction management with rollback capabilities

Model Context Protocol (MCP) Server

• Microservices architecture for tool orchestration
• Real-time tool execution monitoring
• Comprehensive error handling and logging
• Extensible plugin architecture

Interactive Web Interface

• Streamlit-based chat interface with real-time responses
• Visual query result display with data tables
• One-click deployment configuration presets
• System health monitoring and status indicators

Advanced Analytics & Reporting

• Multi-table join operations with complex filtering
• Salary analytics and department performance metrics
• Sales analysis with employee performance tracking
• Real-time schema exploration and relationship mapping

Quick Start Guide

1. Environment Setup

# Clone the repository
git clone <repository-url>
cd oracle-adb-ai-agent

# Dependencies are already installed in the current environment

2. Configuration

Required Environment Variables

# OpenAI Configuration
export OPENAI_API_KEY="sk-your-openai-api-key"

# Oracle ADB Configuration (for production)
export ORACLE_WALLET_LOCATION="/path/to/oracle/wallet"
export ORACLE_WALLET_PASSWORD="your_wallet_password"
export ORACLE_CONNECTION_STRING="service_name_high"
export ORACLE_USERNAME="ADMIN"
export ORACLE_PASSWORD="your_db_password"

# Application Configuration
export LOG_LEVEL="INFO"
export ENVIRONMENT="production"

Alternative: Use Web Interface

• Enter OpenAI API key directly in the web interface sidebar
• Select deployment preset (Development/Production/Testing/Demo)
• System will guide you through configuration

3. Run Application

streamlit run app.py --server.port 5000

Access the application at: http://localhost:5000

Architecture Deep Dive

System Components

┌─────────────────────────────────────────────────────┐
│                Web Interface                        │
│            (Streamlit + Chat UI)                   │
└─────────────────────┬───────────────────────────────┘
                      │
┌─────────────────────▼───────────────────────────────┐
│                MCP Server                           │
│         (Query Orchestration + AI)                 │
└─────────┬───────────────────────────┬───────────────┘
          │                           │
┌─────────▼─────────┐       ┌─────────▼─────────┐
│  Database Tools   │       │   API Tools       │
│ • Query Executor  │       │ • HTTP Requests   │
│ • Schema Explorer │       │ • Authentication  │
│ • Transactions    │       │ • Response Parser │
└─────────┬─────────┘       └─────────┬─────────┘
          │                           │
┌─────────▼─────────┐       ┌─────────▼─────────┐
│   Oracle ADB      │       │  External APIs    │
│ • Employee Data   │       │ • REST Services   │
│ • Department Info │       │ • JSON APIs       │
│ • Order Records   │       │ • Status Checks   │
└───────────────────┘       └───────────────────┘

Data Model

The system includes a comprehensive enterprise data model:

Employees Table

• Employee ID, Name, Department ID
• Salary, Hire Date, Email
• Performance metrics and status

Departments Table

• Department ID, Name, Budget
• Manager ID, Location, Status
• Performance tracking

Orders Table

• Order ID, Customer Name, Amount
• Order Date, Status, Employee ID
• Revenue tracking and analytics

Usage Examples

Basic Database Queries

User: "Show me all employees in the database"
Response: Returns formatted table with employee details

User: "What departments do we have and their budgets?"
Response: Department overview with financial information

User: "Find the top 10 highest paid employees"
Response: Salary rankings with department information

Advanced Analytics

User: "What is the average salary by department?"
Response: Statistical analysis with department comparisons

User: "Show me sales performance by employee for this quarter"
Response: Revenue analysis with individual performance metrics

User: "Find employees hired in the last year working in high-budget departments"
Response: Complex filtered results with multi-table joins

Schema Exploration

User: "What tables are available in the database?"
Response: Complete schema overview with table descriptions

User: "How are employees and departments connected?"
Response: Relationship mapping with foreign key details

User: "Show me the structure of the orders table"
Response: Column details with data types and constraints

API Integration

User: "Test external API connectivity"
Response: API call execution with status and response data

User: "Check system health and tool availability"
Response: Comprehensive system status report

Configuration Management

Deployment Presets

Development Configuration

LOG_LEVEL: DEBUG
API_TIMEOUT: 30 seconds
API_MAX_RETRIES: 3
CACHE_TOOL_RESULTS: enabled
ENVIRONMENT: development

Production Configuration

LOG_LEVEL: INFO
API_TIMEOUT: 15 seconds
API_MAX_RETRIES: 5
CACHE_TOOL_RESULTS: enabled
ENVIRONMENT: production

Testing Configuration

LOG_LEVEL: WARNING
API_TIMEOUT: 60 seconds
API_MAX_RETRIES: 1
CACHE_TOOL_RESULTS: disabled
ENVIRONMENT: testing

Demo Configuration

LOG_LEVEL: INFO
API_TIMEOUT: 20 seconds
API_MAX_RETRIES: 2
CACHE_TOOL_RESULTS: enabled
ENVIRONMENT: demo

File Structure and Responsibilities

oracle-adb-ai-agent/
├── app.py                    # Main Streamlit web interface
├── working_mcp_server.py     # Simplified MCP server implementation
├── mcp_server.py            # Full-featured MCP server
├── database.py              # Database connection and operations
├── oracle_connection.py     # Production Oracle ADB connectivity
├── oracle_tools.py          # Oracle-specific database tools
├── api_tools.py             # External API integration tools
├── config.py                # Configuration management
├── logger.py                # Comprehensive logging system
├── DOCUMENTATION.md         # Complete technical documentation
├── dependencies.txt         # Python package requirements
├── architecture_diagram.svg # System architecture visualization
└── .streamlit/
    └── config.toml          # Streamlit server configuration

Production Deployment

Oracle ADB Setup

1. Download Oracle Wallet

   # From Oracle Cloud Infrastructure Console
   # Navigate to Autonomous Database → DB Connection
   # Download Client Credentials (Wallet)
   

2. Configure Wallet

   # Extract wallet to secure location
   unzip wallet.zip -d /secure/path/wallet/
   
   # Set permissions
   chmod 600 /secure/path/wallet/*
   
   # Set environment variable
   export TNS_ADMIN=/secure/path/wallet/
   

3. Validate Connection

   # Test connection using provided validation script
   python oracle_connection.py
   

Security Best Practices

• API Key Management: Store OpenAI API keys in secure environment variables
• Database Security: Use Oracle Wallet for encrypted authentication
• Network Security: Deploy behind load balancer with SSL/TLS termination
• Access Control: Implement proper user authentication and authorization
• Audit Logging: Enable comprehensive audit trails for all operations

Performance Optimization

• Connection Pooling: Configure optimal pool sizes for concurrent users
• Query Caching: Enable intelligent caching for frequently accessed data
• Resource Monitoring: Implement monitoring for CPU, memory, and database connections
• Load Balancing: Deploy multiple instances behind load balancer for high availability

Troubleshooting Guide

Common Issues and Solutions

Database Connection Problems

Problem: "Oracle connection failed"
Solutions:
1. Verify wallet file permissions (600)
2. Check TNS_ADMIN environment variable
3. Validate connection string format
4. Test network connectivity to Oracle Cloud
5. Verify wallet password correctness

OpenAI API Issues

Problem: "OpenAI API authentication failed"
Solutions:
1. Verify API key format (starts with sk-)
2. Check API key validity on OpenAI platform
3. Verify account billing status and credits
4. Check rate limits and usage quotas
5. Test API connectivity with curl

Tool Execution Failures

Problem: "Tool execution timeout or error"
Solutions:
1. Review application logs for detailed errors
2. Check database connection status
3. Validate query syntax and parameters
4. Verify tool configuration and permissions
5. Test individual tools in isolation

Development and Extension

Adding Custom Tools

1. Create Tool Class

   from langchain.tools import BaseTool
   
   class CustomTool(BaseTool):
       name = "custom_tool"
       description = "Description of tool functionality"
       
       def _run(self, input_data: str) -> str:
           # Implementation logic
           return result
   

2. Register Tool

   # In MCP server initialization
   def _initialize_tools(self):
       self.tools.append(CustomTool())
   

3. Update UI (if needed)

   # Add tool-specific display logic in app.py
   

Custom Database Providers

1. Extend Database Manager

   class CustomDBManager(DatabaseManager):
       def __init__(self, connection_params):
           # Custom initialization
           pass
       
       def execute_query(self, query, parameters):
           # Provider-specific implementation
           pass
   

2. Update Configuration

   # Add provider-specific configuration options
   

API Reference

Core Classes

WorkingMCPServer

• execute_agent_query(query: str) → Dict[str, Any]
• execute_oracle_query(query: str, parameters: Dict) → Dict[str, Any]
• make_api_call(url: str, method: str) → Dict[str, Any]
• check_openai_connection() → bool
• get_database_schema() → Dict[str, Any]

DatabaseManager

• execute_query(query: str, parameters: Dict) → Dict[str, Any]
• begin_transaction() → None
• commit_transaction() → None
• rollback_transaction() → None
• get_schema_info() → Dict[str, Any]

OracleADBConnection

• create_connection_pool(min_conn: int, max_conn: int) → bool
• execute_query(query: str, parameters: Dict, fetch_mode: str) → Dict[str, Any]
• get_schema_info() → Dict[str, Any]
• get_table_details(table_name: str) → Dict[str, Any]
• test_connection() → Dict[str, Any]

Support and Contributing

Enterprise Support

For enterprise deployment assistance, custom integrations, or technical support, contact the development team.

Contributing Guidelines

1. Fork the repository
2. Create feature branch
3. Implement changes with comprehensive tests
4. Update documentation
5. Submit pull request with detailed description

License

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

---

Built with enterprise-grade security, performance, and reliability in mind.