AI Agents
osaurus
Osaurus — own your AI: a native macOS agent harness with any model, memory and fully offline execution.
★ 8.0k starsMIT
View guide →// agents · browser-use/macos-harness
macos-harness
macos-harness is an open-source ai agents project by browser-use (browser-use/macos-harness on GitHub). It has 902 GitHub stars and 61 forks, is released under the MIT license, and was last updated on GitHub on 17 August 2026.
A minimal harness that lets an LLM fully control a Mac.
★ 902 stars⑂ 61 forksMITupdated 2026-08
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
so here's a placeholder. The guide below is complete regardless.
// project facts
Project facts
- Repository
- browser-use/macos-harness
- GitHub stars
- 902
- Forks
- 61
- License
- MIT
- Last updated on GitHub
- 17 August 2026
- Category
- AI Agents
- Listing
- Full install guide
// how it works
How it works
- Clarify your goal
Decide what you want the project to do before installing anything.
- Clone the repo
Grab the source and skim the README for setup requirements.
- Install and configure
Install dependencies and add any API keys or settings it needs.
- Run your first task
Launch it and follow the example workflow from the docs.
// features
Why people love it
Full Mac control
Accessibility APIs drive real apps.
Minimal design
Thin on purpose - easy to understand.
LLM-agnostic
Bring your own model.
Computer-use
OS-level, not just browser-level.
MIT licensed
Free to hack on.
From browser-use team
Built by agent-infra veterans.
Advertisement
// run it
Run it yourself
- Step 1 - Clone the repo
Download the source and move into the project folder.
git clone https://github.com/browser-use/macos-harness.git cd macos-harness - Step 2 - Install dependencies
Create a virtual environment and install the requirements.
python -m venv .venv && source .venv/bin/activate pip install -r requirements.txt - Step 3 - Run an example
Treat these as starting points - the README documents the exact commands.
python examples/quickstart.py
// use cases
What you can build
- Prototype a Mac computer-use agent.
- Automate macOS apps without APIs.
- Research OS-level agent safety.
- Hack on accessibility-driven control.
Advertisement
// faq
Questions, answered
Who maintains macos-harness?
It is an open-source project at browser-use/macos-harness with 61 forks. Check the repo's contributors and recent commits to judge how active maintenance is.
Is macos-harness free to use?
Yes - it is MIT licensed, so you can use, modify, and distribute it freely, including in commercial projects.
Do I need a paid API to use it?
The harness is free; the model driving it is usually a paid API or a local model.
Is this safe to run?
It gives an LLM full control of your Mac - run it in a VM or a spare machine, never on your daily driver with sensitive data. Treat it as research tooling.
How do I get started?
Clone the repo, read the README's quickstart, and run the provided example - most projects get you to a first result in minutes.
// keep exploring
Similar tools
AI Agents
browser-harness
A self-healing harness that connects an LLM directly to your real browser
★ 18.4k starsMIT
View guide →AI Agents
mobilerun
Control Android and iOS devices with natural language — an LLM-agnostic mobile agent framework and CLI.
★ 9.6k starsMIT
View guide →AI Agents
browser-use
The open-source library that lets AI agents control a real browser.
★ 117.4k starsMIT
View guide →AI Agents
MobileBuildMCP
An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.
★ 6.5k starsMIT
View guide →AI Agents
lamda
FIRERPA / LAMDA — full-stack Android device control: remote desktop, automation, MITM, Frida and MCP.
★ 8.6k starsMIT
View guide →// alternatives
Alternatives in AI Agents
The most-starred other AI Agents projects in the LibreAI catalogue — the realistic shortlist if macos-harness is not the right fit.
- openclaw★ 391.6k
A personal AI that really does things — on any OS, with your data staying yours.
- superpowers★ 294.9k
An agentic skills framework & software development methodology that works.
- ECC★ 275.2k
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
- hermes-agent★ 252.0k
The agent that grows with you — a Python framework for adaptive AI agents.
- deepseek-harness★ 245.5k
DeepSeek's plugin-based agent harness where everything is a plugin.
// more in agents
More in AI Agents
Nearby AI Agents projects in the catalogue, closest in stars to macos-harness — keep exploring the shelf either side of it.
- agent-qa★ 902
A self-improving QA agent that writes tests in plain English.
- agentdojo★ 897
Test your agents against prompt-injection attacks.
- motorhead★ 917
A memory and retrieval server built for LLMs.
- SimWorld★ 793
SimWorld: an Unreal Engine 5 simulator for evaluating LLM/VLM agents in physical and social worlds.
- mcp-gateway-registry★ 964
An enterprise MCP gateway: governed, auditable tool access for agents.
- mcp★ 789
An MCP server that puts Hyperbrowser's browsers behind the protocol.
Get macos-harness
Free and open — grab the source, star the repo, and start creating.