OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

MiniAGI

MiniAGI is an open-source Python command-line agent that accepts a user-defined objective and uses a prompt, a small set of tools, reasoning and summarized short-term memory to pursue it. It is compatible with OpenAI GPT-3.5-Turbo and GPT-4, and can use inner monologue and self-criticism.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

MiniAGI — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use MiniAGI when you want to experiment with a command-line agent that combines tool use and summarized short-term memory to pursue a written objective, and can review the commands or code it may produce.

Who it is for

  • Developers who want to explore or adapt a Python-based autonomous agent.
  • Users who want to give an agent a command-line objective involving code or local files.

Documented uses

  • Submit an objective for the agent to pursue using its available tools.
  • Use an objective to generate an artifact, such as a Python-created image or a game.
  • Ask the agent to build and serve a local website using files in its working directory.
Before you choose

Requirements & limits

The repository documents Python 3.10 or higher and use of the OpenAI API with GPT-3.5-Turbo or GPT-4.

  • The critic is enabled by setting ENABLE_CRITIC to true in the environment; the project says it may improve accuracy at the cost of additional API requests.
  • MiniAGI may suggest harmful commands or code, so generated actions need careful review.
  • Depending on settings and requirements, data may be shared with third-party API providers such as OpenAI.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install MiniAGI and prepare its environment

Clone the project, install its requirements and create a local environment file before use.

Before you start

  • Python 3.10 or higher
  • Git
  • Access to the MiniAGI repository

Steps

  1. Clone https://github.com/muellerberndt/mini-agi and change into the mini-agi directory.

  2. Run pip install -r requirements.txt.

  3. Copy .env_example to .env.

Expected outcome

The repository dependencies are installed and a local .env file has been created from the example.

Guide limitations

  • These steps cover setup only; the project’s usage instructions are required to submit an objective.
  • The repository warns that the agent may suggest harmful commands or code and may share data with third-party API providers depending on settings and requirements.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
muellerberndt/mini-agi · Official homepage title · retained in its source language
How we review sourcesSuggest a correction