chatbot

README

OpenRouter.ai Chatbot Example

This example demonstrates how to build a chatbot using the Flow library with OpenRouter.ai integration via the OpenAI Go SDK.

Features

• OpenRouter.ai Integration: Uses OpenRouter.ai API for AI-powered responses
• Conversation History: Maintains conversation context across messages
• Retry Logic: Built-in retry capability using Flow's adaptive node system
• Error Handling: Graceful error handling for API failures

Prerequisites

1. OpenRouter API Key: You need an API key from OpenRouter.ai
2. Go 1.18+: This example requires Go 1.18 or later

Setup

1. Set Environment Variable:

   export OPENROUTER_API_KEY="your-api-key-here"
   

2. Install Dependencies:

   go mod tidy
   

Usage

Run the chatbot:

go run examples/chatbot/main.go

The chatbot will start and you can begin chatting:

OpenRouter.ai Chatbot Example using Flow
Type 'quit' to exit

You: Hello!
Bot: Hello! How can I help you today?

You: What's the weather like?
Bot: I don't have access to real-time weather data, but I'd be happy to help you with other questions!

You: quit
Goodbye!

How It Works

Flow Integration

The chatbot uses Flow's adaptive node system with the following features:

• Retry Logic: Configured with retries: 2 for automatic retry on API failures
• State Management: Uses SharedState to maintain conversation history and client instance
• Parameter-Driven Behavior: Leverages Flow's parameter detection for retry patterns

OpenRouter.ai Configuration

The chatbot is configured to use OpenRouter.ai with:

• Base URL: https://openrouter.ai/api/v1
• Model: openai/gpt-3.5-turbo (can be changed to other OpenRouter models)
• Authentication: Uses the OPENROUTER_API_KEY environment variable

Code Structure

// Initialize OpenAI client with OpenRouter configuration
client := openai.NewClient(
    option.WithAPIKey(apiKey),
    option.WithBaseURL("https://openrouter.ai/api/v1"),
)

// Create Flow node with retry capability
chatNode := flow.NewNode()
chatNode.SetParams(map[string]interface{}{
    "retries": 2,
})

// Set execution function for AI chat completion
chatNode.SetExecFunc(func(prep interface{}) (interface{}, error) {
    // Handle user input and get AI response
    // Maintain conversation history
    // Return response or error
})

Available Models

You can change the model by modifying the Model parameter in the code. Some popular OpenRouter models include:

• openai/gpt-3.5-turbo
• openai/gpt-4
• anthropic/claude-3-haiku
• meta-llama/llama-2-70b-chat
• google/palm-2-chat-bison

Error Handling

The chatbot includes comprehensive error handling:

• API Key Validation: Checks for required environment variable
• Network Errors: Automatic retry with exponential backoff
• API Errors: Graceful error messages for API failures
• Input Validation: Handles empty inputs and quit commands

Customization

You can customize the chatbot by:

1. Changing the Model: Modify the Model parameter to use different AI models
2. Adjusting Retry Logic: Change the retries parameter for different retry behavior
3. Adding System Messages: Include system prompts for specific behavior
4. Extending Features: Add features like conversation saving, user profiles, etc.

Dependencies

• github.com/openai/openai-go: OpenAI Go SDK for API integration
• github.com/joemocha/flow: Flow library for workflow orchestration