Open AI Agent

The OpenAIAgent is a powerful agent class in the Multi-Agent Orchestrator framework that integrates with OpenAI’s Chat Completion API. This agent allows you to leverage OpenAI’s language models, such as GPT-3.5 and GPT-4, for various natural language processing tasks.

Key Features

• Integration with OpenAI’s Chat Completion API
• Support for multiple OpenAI models
• Streaming and non-streaming response options
• Customizable inference configuration
• Handles conversation history for context-aware responses
• Customizable system prompts

Creating an OpenAIAgent

To create a new OpenAIAgent, you need to provide an OpenAIAgentOptions object. Here’s an example of how to create an OpenAIAgent:

Typescript

import { OpenAIAgent } from './path-to-openai-agent';

const agent = new OpenAIAgent({
  name: 'OpenAI Assistant',
  description: 'A versatile AI assistant powered by OpenAI models',
  apiKey: 'your-openai-api-key',
  model: 'gpt-3.5-turbo',
  streaming: true,
  inferenceConfig: {
    maxTokens: 500,
    temperature: 0.7,
    topP: 0.9,
    stopSequences: ['Human:', 'AI:']
  },
  systemPrompt: 'You are a helpful AI assistant specialized in answering questions about technology.'
});

OpenAIAgentOptions

The OpenAIAgentOptions extends the base AgentOptions and includes the following properties:

• name (required): A string representing the name of the agent.
• description (required): A string describing the agent’s capabilities and expertise.
• apiKey (required): Your OpenAI API key.
• model (optional): The OpenAI model to use. Defaults to OPENAI_MODEL_ID_GPT_O_MINI.
• streaming (optional): Whether to use streaming responses. Defaults to false.
• inferenceConfig (optional): An object to customize the inference behavior:
  • maxTokens (optional): The maximum number of tokens to generate. Defaults to 1000.
  • temperature (optional): Controls randomness in output generation.
  • topP (optional): Controls diversity of output generation.
  • stopSequences (optional): An array of sequences that, when generated, will stop the generation process.
• systemPrompt (optional): A string representing the initial system prompt for the agent.

Setting the System Prompt

You can set or update the system prompt for the OpenAIAgent in two ways:

1. During initialization:

Typescript

const agent = new OpenAIAgent({
  // ... other options ...
  systemPrompt: 'You are a helpful AI assistant specialized in answering questions about technology.'
});

2. Using the setSystemPrompt method after initialization:

Typescript

agent.setSystemPrompt(
  `You are an AI assistant specialized in {{DOMAIN}}.
   Your main goal is to {{GOAL}}.
   Always maintain a {{TONE}} tone in your responses.`,
  {
    DOMAIN: "artificial intelligence",
    GOAL: "explain complex AI concepts in simple terms",
    TONE: "friendly and educational"
  }
);

The setSystemPrompt method allows you to dynamically change the agent’s behavior and focus without creating a new instance. You can use placeholders in the prompt template and provide values for them in the second argument.

Usage

Once you’ve created an OpenAIAgent, you can add it to the Multi-Agent Orchestrator and use it to process requests:

Typescript

import { MultiAgentOrchestrator } from "multi-agent-orchestrator";

const orchestrator = new MultiAgentOrchestrator();
orchestrator.addAgent(agent);

const response = await orchestrator.routeRequest(
  "What is the capital of France?",
  "user123",
  "session456"
);

Streaming Responses

If you’ve enabled streaming (streaming: true in the options), the agent will return an AsyncIterable that you can use to process the response in chunks:

Typescript

const streamingResponse = await orchestrator.routeRequest(
  "Tell me a long story about a brave knight",
  "user123",
  "session456"
);

if (Symbol.asyncIterator in streamingResponse) {
  for await (const chunk of streamingResponse) {
    console.log(chunk); // Process each chunk of the response
  }
}

Best Practices

1. API Key Security: Ensure your OpenAI API key is kept secure and not exposed in your codebase.
2. Model Selection: Choose an appropriate model based on your use case and performance requirements.
3. Inference Configuration: Experiment with different inference parameters to find the best balance between response quality and speed.
4. Error Handling: Implement additional error handling in your application to manage potential API failures gracefully.
5. Rate Limiting: Be aware of OpenAI’s rate limits and implement appropriate throttling if necessary.
6. System Prompts: Craft clear and specific system prompts to guide the model’s behavior and improve response quality for your use case.

By leveraging the OpenAIAgent, you can create sophisticated, context-aware AI agents capable of handling a wide range of tasks and interactions, all powered by OpenAI’s state-of-the-art language models.