GitHub logo

GitHub

Free

MiniAGI is a simple general-purpose AI agent based on the OpenAI API.

FreeFree tier
Inputs: textOutputs: code, text
Type
Open Source

About GitHub

MiniAGI is a simple general-purpose autonomous AI agent based on the OpenAI API. It supports GPT-3.5-Turbo and GPT-4, combining a robust prompt with a minimal set of tools, chain-of-thought reasoning, short-term memory with summarization, inner monologue, and self-criticism. The agent can execute Python code and shell commands to accomplish user-defined objectives autonomously. It is open-source and designed for lightweight, flexible automation.

Key Features

Supports GPT-3.5-Turbo and GPT-4
Chain-of-thought reasoning and inner monologue
Short-term memory with summarization
Self-criticism capability for improved accuracy
Can execute Python code and shell commands
Minimal and lightweight design
Autonomous objective completion

Pros & Cons

Pros
  • Open source and free to use (no subscription)
  • Lightweight and easy to install
  • Supports both GPT-3.5 and GPT-4 models
  • Self-criticism enhances output accuracy
  • Capable of executing Python and shell commands autonomously
Cons
  • Requires OpenAI API key and incurs API costs
  • May inadvertently suggest harmful commands (safety advisory)
  • Critic mode increases API usage and cost
  • Limited to OpenAI models only
  • No built-in web search or external API integration

Best For

Programmatically draw and save imagesCreate and serve websitesGenerate online income ideas autonomouslyGeneral-purpose task automationFile manipulation and shell scripting

FAQ

What is MiniAGI?
MiniAGI is a simple autonomous AI agent based on the OpenAI API. It uses chain-of-thought, short-term memory, and self-criticism to accomplish given objectives.
How do I install MiniAGI?
Clone the repository, run 'pip install -r requirements.txt', copy '.env_example' to '.env', and set your OpenAI API key.
What models does MiniAGI support?
MiniAGI is compatible with GPT-3.5-Turbo and GPT-4.
How do I enable the critic?
Set 'ENABLE_CRITIC' to 'true' in your .env file. This may improve accuracy at the cost of additional API requests.
Is MiniAGI safe to use?
The repository includes a safety advisory: the agent may inadvertently suggest harmful commands or code. Proceed with caution and use at your own discretion.