# muellerberndt/mini-agi

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

Repository: https://github.com/muellerberndt/mini-agi
Canonical: https://ross.abutalabs.com/products/mini-agi
Language: Python
License: MIT
License Family: permissive
Topics: agent, agi, artificial-intelligence, gpt-3, gpt-4
Last push: 2023-06-14T05:47:59+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 88
- inputs: {"age_days": 1237, "days_push": 1176, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2899, forks 307 (observed 2026-08-28T04:07:28.831584+00:00)

## What it is
MiniAGI is a simple general-purpose autonomous AI agent built on the OpenAI API, compatible with GPT-3.5-Turbo and GPT-4. It combines a robust prompt with a minimal set of tools, chain-of-thought reasoning, short-term memory with summarization, inner monologue, and optional self-criticism.

## Use cases
- run an autonomous AI agent to complete objectives
- have GPT-4 execute Python code and shell commands automatically
- experiment with minimal agent architectures and prompting
- build a website or draw images from a natural language objective
- explore chain-of-thought reasoning with short-term memory summarization
- study self-criticism and inner monologue in LLM agents

## When to choose
- you want a minimal, readable codebase to learn how autonomous agents work
- you need a lightweight GPT-3.5/GPT-4 agent with a small toolset
- you want to experiment with agent prompting techniques like inner monologue and self-criticism

## When to avoid
- you need a production-grade agent framework with extensive integrations and plugins
- you cannot send data to third-party API providers like OpenAI
- you need an actively developed project with recent updates
- you require strict safety guarantees, since the agent may suggest harmful commands or code

## Facets
- artifact type: application
- maturity: maintenance
- function: agent-framework, llm-inference, prompt-engineering
- domain: artificial-intelligence, large-language-models
- platform: python, cli, cross-platform
- tags: autonomous-agent, gpt-4, gpt-3, openai-api, agi, chain-of-thought, short-term-memory, self-criticism, ai-agents, automation

## Member repositories
- muellerberndt/mini-agi (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.831584+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:35:11.660166+00:00, confidence not recorded.
  - readme: https://github.com/muellerberndt/mini-agi (fetched 2026-08-28T04:07:28.831584+00:00, sha fccd06b00c8f)
- Data as of 2026-08-30T08:39:29.467469+00:00.