AI

📸 Screenshots

Here are visual examples of the AI worker:

AI Worker Interface AI Worker - AI model integration and processing

The Ai Worker is a bot designed to interact with various AI models to generate text based on given prompts and documents. It is capable of working with different AI providers such as OpenAI, Anthropic, Google Generative AI, DeepSeek, Groq, and Xai. The worker is designed to be flexible and adaptable, allowing for easy integration with different AI models and providers.

2. Configuration Parameters

The Ai Worker has two configurable parameters:

temperature: This parameter controls the randomness of the AI's output. A higher value results in more random outputs, while a lower value results in more deterministic outputs.
model: This parameter specifies the AI model to be used. It should be in the format provider/modelID, where provider is the name of the AI provider (e.g., "openai") and modelID is the ID of the specific model to be used.

3. Input/Output Handles

The Ai Worker has five input/output handles:

prompt: This input handle accepts a string that serves as the prompt for the AI model.
input: This input handle accepts a string that serves as additional input for the AI model.
documents: This input handle accepts a document that the AI model can use for reference.
history: This input handle accepts a chat history that the AI model can use for context.
answer: This output handle outputs the generated text from the AI model.

4. Usage Examples with Code

Here's an example of how to use the Ai Worker:

const worker = create(agent)
await worker.execute({
  prompt: "What is the meaning of life?",
  input: "42",
  documents: ["The Hitchhiker's Guide to the Galaxy"],
  history: [
    { role: "user", content: "What is the meaning of life?" },
    { role: "assistant", content: "42" },
  ],
  answer: "",
})

5. Integration Examples

The Ai Worker can be integrated into a chatbot application to provide intelligent responses to user queries. Here's an example:

const chatbot = new Chatbot(agent)
chatbot.onMessage(async (message) => {
  const worker = create(agent)
  const response = await worker.execute({
    prompt: message.content,
    input: "",
    documents: [],
    history: chatbot.history,
    answer: "",
  })
  chatbot.sendMessage(response.answer)
})

6. Best Practices

Always provide as much context as possible to the AI model through the prompt, input, documents, and history handles. This will help the model generate more accurate and relevant responses.
Be sure to handle errors properly. The Ai Worker will set the error property if something goes wrong, so always check this property after calling execute().

7. Troubleshooting Tips

If the Ai Worker is not generating any output, make sure that you have provided a valid prompt and input, and that the model parameter is correctly set.
If you're getting an error message saying "No model selected" or "No API key found", make sure that you have correctly set the model parameter and that you have provided a valid API key for the specified provider.

AI