Customer Support Bot

A conversational AI customer support bot built with LangGraph, Ollama, and LangChain. The bot can handle customer inquiries and use tools to provide helpful information.

Features

• Conversational AI: Powered by Llama 3.2 model via Ollama
• Tool Integration: Built-in tools for common support tasks
• Memory: Maintains conversation context across interactions
• Extensible: Easy to add new tools and capabilities

Available Tools

1. Current Time: Get the current date and time
2. Random Number Generator: Generate random numbers within a specified range
3. Calculator: Perform basic mathematical operations (addition)
4. Customer Support Info: Provide information about billing, technical support, returns, shipping, and account management

Prerequisites

• Python 3.8 or higher
• Windows, macOS, or Linux

Installation & Setup

1. Clone or Download the Project

git clone <your-repository-url>
cd "customer support bot"

2. Create a Virtual Environment

# Create virtual environment
python -m venv venv

# Activate virtual environment
# On Windows:
venv\Scripts\activate

# On macOS/Linux:
source venv/bin/activate

3. Install Python Dependencies

pip install -r requirements.txt

4. Install Ollama

Windows:

1. Download Ollama from https://ollama.ai/
2. Run the installer and follow the setup instructions

macOS:

brew install ollama

Linux:

curl -fsSL https://ollama.ai/install.sh | sh

5. Start Ollama Service

Open a new terminal window and start the Ollama service:

ollama serve

Keep this terminal window open while using the bot.

6. Download the Llama 3.2 Model

In another terminal window, download the required model:

ollama pull llama3.2

This will download the Llama 3.2 model (approximately 2GB). Wait for the download to complete.

7. Verify Ollama Installation

Check if the model is available:

ollama list

You should see llama3.2 in the list of available models.

Usage

Running the Bot

1. Make sure Ollama is running (ollama serve in a separate terminal)
2. Activate your virtual environment
3. Run the bot:

python flow.py

Example Interactions

Getting Current Time:

🤖 Pass your prompt here: What time is it?

Generating Random Numbers:

🤖 Pass your prompt here: Generate a random number between 1 and 100

Mathematical Calculations:

🤖 Pass your prompt here: What's 25 plus 37?

Customer Support:

🤖 Pass your prompt here: I need help with billing
🤖 Pass your prompt here: How do I return an item?
🤖 Pass your prompt here: What are your shipping options?

General Conversation:

🤖 Pass your prompt here: Hello, how can you help me?

Exiting the Bot

Type quit, exit, or press Ctrl+C to stop the bot.

Project Structure

customer support bot/
├── flow.py              # Main application file
├── tool.py              # Tool definitions
├── requirements.txt     # Python dependencies
├── .gitignore          # Git ignore file
└── README.md           # This file

Troubleshooting

Common Issues

1. Connection Error: If you see a connection error, make sure Ollama is running with ollama serve

2. Model Not Found: If you get a model error, ensure you've downloaded the model with ollama pull llama3.2

3. Import Errors: Make sure you've activated your virtual environment and installed all requirements

4. Tool Binding Error: Ensure you're using the correct Ollama model that supports tool calling

Checking Ollama Status

# Check if Ollama is running
curl http://localhost:11434/api/tags

# List available models
ollama list

# Test model directly
ollama run llama3.2 "Hello"

Development

Adding New Tools

To add new tools, edit tool.py:

1. Create a new function with the @tool decorator
2. Add proper type hints and docstring
3. Add the tool to the X list at the bottom of the file

Example:

@tool
def my_new_tool(param: str) -> str:
    """Description of what this tool does."""
    return f"Result: {param}"

Customizing the Bot

• Change Model: Edit the model name in flow.py (line 12)
• Modify Personality: Add system prompts or modify the chatbot function
• Add Memory: The bot already has conversation memory enabled

Dependencies

• langgraph: For building conversational AI workflows
• langchain-ollama: For integrating with Ollama models
• langchain-core: Core LangChain functionality
• colorama: For terminal text coloring
• typing-extensions: Enhanced type hints
• pydantic: Data validation

License

This project is open source. Feel free to modify and distribute as needed.

Support

For issues or questions:

1. Check the troubleshooting section above
2. Ensure all installation steps were completed
3. Verify Ollama is running and the model is downloaded