Open Interpreter

Open Interpreter is a coding agent that lives in your terminal. Open a project, type interpreter, and it reads your files, edits them, and runs commands on your behalf.

It is built on top of Codex and stays provider agnostic, so you can plug in OpenAI, Anthropic, a local model, or anything else.

Install

Get the interpreter command on your machine.

Quickstart

Run your first session in under a minute.

Authenticate

Configure

Tune providers, sandbox, and features.

What it does

Edits code in place. Ask for a change and review the diff before it lands.
Runs commands with explicit guardrails. A sandbox decides what is even possible, and an approval mode decides when to ask first.
Brings your own model. Pick the provider and model that fit the task. Switch any time with /model.
Resumes long sessions. Pick up earlier work with interpreter resume --last.

How it feels

$ cd my-project
$ interpreter
> add a /health endpoint that returns the build sha

Open Interpreter plans the change, edits the files, and asks before it escalates beyond the current sandbox.

Where to go next

Slash commands

Switch models, manage sessions, and more from the composer.

AGENTS.md

Give the agent project-specific guidance it always reads.

Skills

Package reusable workflows the agent can pick up automatically.

Sandbox & approvals

Decide what runs automatically and what asks first.

Edit on GitHub