AgentGPT

What is AgentGPT ?

AgentGPT allows you to configure and deploy Autonomous AI agents. Name your own custom AI and have it embark on any goal imaginable. It will attempt to reach the goal by thinking of tasks to do, executing them, and learning from the results .

AgentGPT Demo

For the best demo experience, try the site

AgentGPT Features

• ✅ Bootstrapping: create-t3-app + FastAPI-template.
• ✅ Framework: Nextjs 13 + Typescript + FastAPI
• ✅ Auth: Next-Auth.js
• ✅ ORM: Prisma & SQLModel.
• ✅ Database: Planetscale.
• ✅ Styling: TailwindCSS + HeadlessUI.
• ✅ Schema Validation: Zod + Pydantic.
• ✅ LLM Tooling: Langchain. This platform is currently in beta, a full list of completed and planed features can be found on the public roadmap.

Install AgentGPT

The easiest way to get started with AgentGPT is automatic setup CLI bundled with the project. The cli sets up the following for AgentGPT:

• 🔐 Environment variables (and API Keys)
• 🗂️ Database (Mysql)
• 🤖 Backend (FastAPI)
• 🎨 Frontend (Nextjs)

Prerequisites

Before you get started, please make sure you have the following installed:

• An editor of your choice. For example, Visual Studio Code (VS Code)
• Node.js
• Git
• Docker. After installation, please create an account, open up the Docker application, and sign in.
• An OpenAI API key
• A Serper API Key (optional)
• A Replicate API Token (optional)

Getting Started

1. Open your editor
2. Open the Terminal - Typically, you can do this from a ‘Terminal’ tab or by using a shortcut (e.g., Ctrl + ~ for Windows or Control + ~ for Mac in VS Code).
3. Clone the Repository and Navigate into the Directory - Once your terminal is open, you can clone the repository and move into the directory by running the commands below. For Mac/Linux users :apple: :penguin:

   git clone https://github.com/reworkd/AgentGPT.git
   cd AgentGPT
   ./setup.sh

For Windows users :windows:

   git clone https://github.com/reworkd/AgentGPT.git
   cd AgentGPT
   ./setup.bat

1. Follow the setup instructions from the script - add the appropriate API keys, and once all of the services are running, travel to http://localhost:3000 on your web-browser.