Skip to main content

Build Your Own Agent

Napthaville Tutorial

Build your own multi-agent simulation inspired by "Smallville" research project.

Project Overview​

This guide walks through creating a small town simulation using Naptha's agent framework Github. You'll learn how to:

  • Create an Agent by defining its unique persona.
  • Submit a pull request on GitHub to become a citizen.
  • Run interactive simulations with generative agents.
  • Build your own world by customizing our example.

Based on the "Generative Agents: Interactive Simulacra of Human Behavior" research paper.

Step 1: Create Your Own Agent​

An agent represents an entity that can be configured to respond to a user's messages using several parameters like model, instructions, and tools.

Example from the repository:

{
  "name": "Abigail Chen",
  "maze": "the_ville",
  "location": {
    "x": 36,
    "y": 18
  }
}

Step 2: Submit a Pull Request​

A Building represents a conversation between a user and one or many Assistants. You can create a Thread when a user (or your AI application) starts a conversation with your Assistant.

Step 3: Run Simulations​

The contents of the messages your users or applications create are added as Message objects to the Thread. Messages can contain both text and files.

Step 4: Build a New World​

Once all the user Messages have been added to the Thread, you can 'Run the Thread' with any Assistant. Creating a Run uses the model and tools associated with the Assistant to generate a response. These responses are added to the Thread as Assistant Messages.

What’s next?​