# OthersideAI/self-operating-computer

A framework to enable a multimodal model to operate a computer.

Repository: https://github.com/OthersideAI/self-operating-computer
Canonical: https://ross.abutalabs.com/products/self-operating-computer
Homepage: https://www.hyperwriteai.com/self-operating-computer
Language: Python
License: MIT
License Family: permissive
Topics: automation, openai, pyautogui
Last push: 2025-09-19T19:31:39+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 42, release rhythm 40, longevity 73
- inputs: {"age_days": 1033, "days_push": 348, "days_rel": 551, "gap_med": 0, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10291, forks 1421 (observed 2026-08-28T04:10:41.529629+00:00)

## What it is
A Python framework that lets multimodal LLMs operate a computer by viewing the screen and issuing mouse and keyboard actions to reach a stated objective. It integrates with models like GPT-4o, Gemini Pro Vision, Claude 3, and Qwen-VL and is driven via the `operate` CLI.

## Use cases
- automate computer tasks with an ai agent
- let an llm control mouse and keyboard
- computer-use agent framework
- automate gui interactions with gpt-4o
- ai that operates a desktop like a human
- run multimodal model to complete on-screen objectives

## When to choose
- you want an open-source computer-use agent that works with multiple multimodal model providers
- you need to automate desktop GUI tasks via screen understanding
- you want a pip-installable CLI for AI-driven OS automation

## When to avoid
- you need only browser automation rather than full desktop control
- you cannot grant screen-recording and accessibility permissions to your terminal
- you require a fully sandboxed or headless automation environment

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, workflow-automation, computer-vision, llm-inference, cli
- domain: artificial-intelligence, developer-tools
- platform: python, windows, cli
- tags: computer-use, multimodal-models, screen-control, pyautogui, gui-automation, ai-agents, automation, macos, linux

## Member repositories
- OthersideAI/self-operating-computer (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:41.529629+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 2026-08-29T17:19:12.422380+00:00, confidence not recorded.
  - readme: https://github.com/OthersideAI/self-operating-computer (fetched 2026-08-28T04:10:41.529629+00:00, sha d64ad4f2b617)
- Data as of 2026-08-30T08:39:29.467469+00:00.