Overview
Open Interpreter is an open-source coding agent that runs in your terminal and executes code directly on your own machine to carry out tasks you describe in plain language. This is the Rust version of the project, built as a fork of OpenAI's Codex with a focus on emulating the agent harness that gets the best results from low-cost models.
It is aimed at developers who want a local agent rather than a hosted one: commands run inside native sandboxing on macOS, Linux, and Windows, and config and session state stay local under ~/.openinterpreter. You can switch between providers and models with /model and swap the underlying agent harness with /harness.
As a computer-use agent framework, it goes beyond running shell commands. A built-in QA skill lets a model operate and test interfaces, driving web apps in a real browser through agent-browser or operating native apps through trycua.
What it does
- Runs commands inside native sandboxing on macOS, Linux, and Windows
- Switches providers and models from the TUI with /model
- Inspects or switches Rust-native model harnesses with /harness (native, claude-code, kimi-cli, qwen-code, and more)
- Tests web and native apps through a built-in QA skill, via agent-browser or trycua
- Shares one local runtime across terminal tabs instead of starting a full runtime per session
- Supports exec, MCP, skills, hooks, permissions, and AGENTS.md; keeps config under ~/.openinterpreter
Getting started
Install with the official script for your OS, then start a session from the terminal.
Install on macOS or Linux
Run the official install script. After it finishes, the i and interpreter commands are available in your terminal.
curl -fsSL https://openinterpreter.com/install | shInstall on Windows
Use the PowerShell install script instead.
irm https://openinterpreter.com/install.ps1 | iexStart a session
Type i or interpreter in your terminal to open an interactive session.
interpreterSwitch the active harness
Inside a session, use /harness to inspect or switch the model harness, and /model to switch providers and models.
/harnessCommands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Automate local development tasks from plain-language requests, such as running scripts, editing files, or fixing errors
- Operate low-cost or self-hosted models through a harness tuned for agent performance
- Drive and QA-test a web app in a real browser without writing the automation by hand
- Operate and test native desktop apps through the built-in computer-use skill
How Open Interpreter compares
Open Interpreter alongside other open-source computer & browser use tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| Browser Use | ★ 105k | A Python library that lets agents control a real browser to read pages and complete tasks online from natural-language instructions. |
| Open Interpreter | ★ 65.9k | A terminal coding agent that runs code on your own computer from natural-language requests |
| UI-TARS Desktop | ★ 38k | ByteDance's multimodal agent stack and desktop app that controls a computer's graphical interface using vision-language models. |
| Stagehand | ★ 23.5k | A TypeScript browser-automation SDK from Browserbase that mixes natural-language actions with normal code for reliable web agents. |
| Skyvern | ★ 22.4k | A tool that uses language models and computer vision to automate browser workflows without writing custom code for each website. |
| Cua | ★ 19.9k | Infrastructure for computer-use agents, providing sandboxes and SDKs that let agents control full desktops on macOS, Linux, and Windows. |
| Agent Zero | ★ 18.4k | A general-purpose personal agent framework that uses the computer, terminal, and code as its main tools to complete tasks. |
| Browser Use Web UI | ★ 16.2k | A user-friendly web UI, built on Gradio, that lets you control browser-use AI agents, pick from many LLMs, and use your own browser. |