LangGraph Agentic Workflow with Streamlit Frontend

This project combines LangGraph for advanced agentic AI workflows with a Streamlit frontend to create an interactive chatbot application. The system features persistent conversation threads stored in a SQLite database, uses Mistral AI for language processing, and includes multiple agentic tools for enhanced functionality.

Features

• Multi-Tool Agentic AI Workflow: Powered by LangGraph with support for multiple tools including web search, stock price lookup, and calculator
• Interactive Streamlit Interface: User-friendly chat interface with conversation management
• Persistent Conversations: Multiple chat threads stored in SQLite database
• Thread Management: Create new chats and switch between existing conversations
• Real-time Streaming Responses: AI responses streamed directly to the UI
• Web Search Capability: DuckDuckGo integration for current information retrieval
• Stock Price Lookup: Real-time stock price fetching using Alpha Vantage API
• Calculator Tool: Perform basic mathematical operations directly in chat

Architecture

Backend (LangGraph)

• State management with LangGraph's StateGraph
• Multi-tool orchestration with conditional routing
• Conversation history stored in SQLite database using SqliteSaver
• Mistral AI integration for language processing
• Thread-safe checkpointing system
• Tool integration including web search, stock prices, and calculator

Frontend (Streamlit)

• Real-time chat interface
• Conversation sidebar for managing multiple threads
• Session state management
• Message streaming capability
• Thread switching functionality

Prerequisites

• Python 3.8 or higher
• Mistral AI API key
• Alpha Vantage API key (for stock price functionality)

Installation

1. Clone this repository:

   git clone <repository-url>
   cd <repository-name>

2. Install the required dependencies:

   pip install -r requirements.txt

3. Set up your environment variables by creating a .env file:

   MISTRAL_API_KEY=your_mistral_api_key_here
   ALPHA_VANTAGE_API_KEY=your_alpha_vantage_api_key_here

Usage

1. Run the Streamlit application:

   streamlit run Streamlit_frontend_database.py

2. The application will open in your default browser at http://localhost:8501

3. Start a new conversation by typing in the chat input, or select from existing conversations in the sidebar

Project Structure

• Streamlit_frontend_database.py - Streamlit frontend with UI components
• langraph_database.py - LangGraph workflow definition, database integration, and tool implementations
• requirements.txt - Python dependencies
• chat.db - SQLite database for conversation persistence

Configuration

The application uses the following configuration:

• AI Model: Mistral Small (configured in langraph_database.py)
• Database: SQLite (chat.db)
• APIs: Mistral AI and Alpha Vantage (requires API keys in .env)
• Tools: DuckDuckGo Search, Stock Price Lookup, and Calculator

How It Works

1. Conversation Threads: Each chat session gets a unique thread ID stored in the database
2. Message Flow: User messages → LangGraph workflow → Mistral AI → Tool selection if needed → Response streaming
3. Tool Integration: If tools are detected, LangGraph conditionally routes to tool execution nodes
4. State Management: Conversation history is maintained between interactions
5. Persistence: All conversations are saved to SQLite and can be retrieved
6. Multi-step Workflows: Complex queries can trigger multiple tool calls with intelligent routing

Available Tools

The agentic workflow includes these integrated tools:

• DuckDuckGo Search: Search the web for current information
• Stock Price Lookup: Retrieve real-time stock prices using Alpha Vantage API
• Calculator: Perform basic mathematical operations (add, subtract, multiply, divide)

Customization

You can modify the following aspects:

• AI Model: Change the model in langraph_database.py
• Prompt Engineering: Modify the question_ans function logic
• UI Elements: Customize the Streamlit interface in Streamlit_frontend_database.py
• State Schema: Extend the ChatState in langraph_database.py
• Add New Tools: Create additional tools by implementing the @tool decorator pattern
• Tool Logic: Modify existing tool implementations in langraph_database.py

Troubleshooting

• API Key Issues: Ensure your MISTRAL_API_KEY and ALPHA_VANTAGE_API_KEY are correctly set in the .env file
• Database Issues: Check that the chat.db file has proper read/write permissions
• Tool Issues: Verify that all required API keys are properly configured for tool functionality
• Streamlit Issues: Make sure all dependencies are installed properly

Security Note

Never commit your API keys to version control. The .env file should be in your .gitignore file.