Getting started

Install loop, sign in to a model provider, and run your first session.

Install it, give it a model, and start talking to it. Two commands and about a minute.

Install

macOS, Linux, or WSL:

curl -fsSL https://raw.githubusercontent.com/notshekhar/loop/main/install.sh | bash

Windows PowerShell:

irm https://raw.githubusercontent.com/notshekhar/loop/main/install.ps1 | iex

These drop a prebuilt binary — no Node, no build step. Full details, other platforms, and updating are in Installing.

loop has no models of its own. It runs whatever you point it at: a subscription you already pay for, an API key, or a model running locally on your machine.

loop login

That opens a provider picker. The three cheapest ways to start:

Already pay for SuperGrok? loop login xai → pick OAuth. Browser sign-in, no key to paste, billed to your plan.
Already pay for GitHub Copilot or ChatGPT? loop login github-copilot or loop login openai → pick the subscription option.
Want to pay nothing? Run Ollama locally. loop detects the daemon and lists your local models with no login at all.

Otherwise paste an API key. Signing in has the console URL for every provider and what each key is called.

Run it

cd ~/code/your-project
loop

That's the TUI. Type what you want in plain English; loop reads files, edits them, runs commands, and shows you diffs as it goes.

loop run "explain what this repo does"

That's one-shot mode — a single prompt, printed to stdout, no UI. Good for scripts and CI.

The first five minutes

Do this	What happens
`/help`	Every slash command, searchable
`/model`	Switch model — any provider you're signed in to
`/init`	loop reads the repo and writes an `AGENTS.md` so it knows your conventions
`Tab` on an empty prompt	Cycle agents: default → plan (read-only) → data-analyst
`Esc`	Interrupt the turn in progress
`/tree`	Navigate the session as a tree and branch from any earlier message

Messages typed while the agent is working are queued and run when the turn ends — you don't have to wait for a prompt.

Where to go next

Installing — every platform, updating, building from source
Signing in — every provider, where to get each API key, custom gateways
Daily use — sessions, the session tree, models, agents
Telegram — drive loop from your phone
Automation — background tasks, goals, scheduled runs
Extending — MCP servers, hooks, custom agents, extensions
Configuration — what lives in ~/.loop, settings, adding models
Troubleshooting — when something doesn't work

What loop actually is

An open-source terminal coding agent. One TUI in front of 15+ model providers, with a few things that aren't standard:

Sessions are trees, not transcripts. Branch from any earlier message and keep both paths. Abandoned branches get summarized back into context instead of thrown away.
Background tasks run detached, on an OS timer (launchd, systemd, Task Scheduler), so they fire even when loop is closed.
Goal mode drives a session autonomously until an adversarial verifier agrees the objective is met.
Claude Code-compatible hooks — your existing ~/.claude hooks and plugins load as-is.
Remote gateways — pair a Telegram bot and drive the agent from your phone.