# Omarchy for AI Coding: Claude Code, Codex, Cursor and Agent Setup (2026)

> How to set up Omarchy 4.5 for AI coding: built-in agent launchers, Claude Code and Codex, multiple subscriptions, tmux agent layouts, worktrees, notifications, hotkeys and the gotchas to know.

- URL: https://acceptallhq.com/guides/omarchy-ai-coding/
- Section: Guides
- Published: 2026-10-01T17:34:08-04:00
- Updated: 2026-11-05T12:00:00-05:00
- Author: MK (https://acceptallhq.com/about/)
- Verified against: Omarchy 4.5.0 (released Nov 5, 2026), the latest stable release, checked against the official manual, the 4.5 release notes, the omacom/omarchy releases and the 4.5.0 source on November 5, 2026.

**TL;DR:** Omarchy 4.5 ("Quattro RS") is the most agent-ready Linux desktop you can install today. It ships lazy launchers for two dozen coding agents, a one-key agent window, a drop-down console that starts your default agent, several subscriptions per agent with automatic switching, tmux layouts built for running several agents at once, and a top-bar panel for your plan limits. This guide covers what's already there, what to install, and the auto-approve defaults you should know about before you let an agent loose.

## What Omarchy is (in one paragraph)

Omarchy is DHH's opinionated Arch Linux + Hyprland setup, now pitched as "the malleable OS for the age of agents." Version 4 rewrote the desktop shell in Quickshell, moved Omarchy itself into real Arch packages, and made Foot the default terminal and Neovim the default editor. Version 4.5 (Nov 5, 2026) adds a traditional desktop mode next to the tiling default and native installs on Apple Silicon MacBooks. On a PC you install it from an ISO (flash it, boot it, done in a few minutes). Secure Boot has to be off, and because the disk is encrypted by default, type the unlock password on a wired or 2.4 GHz keyboard, not Bluetooth.

## The AI stuff that's already installed

This is the part most "install Claude Code on Linux" tutorials miss: on Omarchy 4 you mostly don't install agents, you just run them.

- **Lazy agent launchers.** In 4.5.0, `omarchy-default-agent` knows 24 agents: `claude`, `codex`, `opencode`, `agy` (Google's Antigravity CLI), `amp`, `commandcode`, `copilot`, `crush`, `cursor-agent`, `devin`, `dsh`, `freebuff`, `grok`, `hermes`, `kilo`, `kimi`, `kiro-cli`, `mcode`, `muse`, `omp`, `openclaw`, `ori`, `pi` and `qwen`. Nothing downloads until the first time you run one (most install through mise or npm), and `omarchy update` keeps them current. Hermes is the exception: picking it installs Hermes Desktop. Antigravity replaced the obsolete Gemini CLI in 4.5; `gemini` still works as an alias for `agy`.
- **A default agent.** Pick one with `omarchy default agent <name>` (or *Setup > Defaults > Agent* in the menu). After that, `Super+Shift+Ctrl+A` opens it in its own window, `Super+Grave` drops a Quake-style console over the desktop that starts the default agent the first time you open it, and `omarchy agent prompt "review this repo"` starts it straight into a task. Launches from your home folder land in `~/Work` instead.
- **Short aliases.** `a` runs your default agent in the current terminal, `cx` runs Claude Code, `cy` runs Codex, `c` runs opencode.
- **Several subscriptions per agent.** New in 4.5: `omarchy agent account add claude Work` keeps a second Claude (or Codex or Grok) login, `omarchy agent account use <name>` switches, and `omarchy agent account mode auto` moves new sessions to the account with the most headroom when one nears its limit. History, settings and skills are shared, so `claude --continue` picks up after a switch. `omarchy agent account copy user@host` sends logins to another machine over SSH, so only to one you trust.
- **A usage panel.** The agents icon in the top bar shows the plan and limits for each signed-in account, with a meter per limit and the time until it resets, refreshed every 15 minutes. 4.5 grew it from 6 to 23 agents and providers, Claude Code, Codex, Grok, Cursor, Copilot, Antigravity and Ollama among them. Left-click to check, right-click to launch.
- **An Omarchy skill for your agents.** It's linked into Claude Code's and Codex's skill folders so an agent can tweak your desktop config. The manual calls it experimental: use plan mode first.

**Read this before you type `cx`:** the aliases and the one-key launcher start agents in auto-approve modes. In the 4.5.0 source, `cx` is `claude --permission-mode auto`, `cy` is `codex --approve-for-me`, `c` is `opencode --auto`, and the launcher goes further for others: `cursor-agent --yolo --trust`, `copilot --allow-all`, `crush --yolo`, `agy --dangerously-skip-permissions`. The manual's own warning is that agents launched this way "run unattended." That's the point, and it's also how every story in our [Agent Fail Hall of Fame](/agent-fails/) starts. Only run them inside a project folder or a git worktree, never at the root of your home directory.

## Installing the rest

Everything below was checked against each vendor's official install docs or the Arch package pages.

| Tool | On Omarchy | Plain Arch / official route |
|---|---|---|
| Claude Code | Already stubbed: run `claude` | `curl -fsSL https://claude.ai/install.sh \| bash` |
| OpenAI Codex CLI | Already stubbed: run `codex` | `sudo pacman -S openai-codex` (Arch extra) |
| Antigravity CLI (Google's replacement for Gemini CLI) | Already stubbed: run `agy` (`gemini` is kept as an alias for it) | Download from antigravity.google. `sudo pacman -S gemini-cli` still installs, but Google stopped serving Gemini CLI to Google-account logins on June 18, 2026; it now needs a paid Gemini API key |
| opencode | Already stubbed: run `opencode` | `sudo pacman -S opencode` (Arch extra) |
| Cursor | CLI already stubbed: run `cursor-agent`. Editor via *Install > Editor > Cursor* (AUR package `cursor-bin`) | AppImage from cursor.com; CLI via `curl https://cursor.com/install -fsS \| bash` |
| VS Code + Copilot | *Install > Editor > VSCode* (AUR `visual-studio-code-bin`, themed, keyring pre-set) | Then "Set up Copilot" from the status bar |
| Ollama (local models) | *Install > AI > Ollama* (picks the CUDA or ROCm build for your GPU) | `sudo pacman -S ollama` (or `ollama-cuda` / `ollama-rocm`) |
| LM Studio | *Install > AI > LM Studio* | The manual's pick for local-model beginners |

Two notes. First, don't run `sudo pacman -Syu` or `yay -Syu` to update: Omarchy blocks full system upgrades on purpose and wants `omarchy update` instead, which snapshots first so you can roll back from the boot menu. Installing single packages with `pacman -S` is fine. Second, if an agent decides to "update the system" for you, it will hit that same block. Good.

## Hotkeys worth learning first

From the Omarchy manual's hotkey page. `Super+K` shows the full list any time.

| Keys | What it does |
|---|---|
| `Super+Space` | Omarchy menu |
| `Super+Return` | Terminal |
| `Super+Alt+Return` | tmux |
| `Super+Ctrl+Return` | Herdr, the bundled agent manager (`Super+Ctrl+K` lists its bindings) |
| `Super+Shift+Ctrl+A` | Your default AI agent |
| `Super+Grave` | Quake console (starts your default agent on first open; `Super+Shift+Grave` sends a window there) |
| `Super+1` to `4` | Switch workspace |
| `Super+Shift+Alt+1` to `4` | Send a window to a workspace without following it |
| `Super+L` | Cycle the workspace layout: dwindle, scrolling (great for lots of agent terminals), floating. `Super+Alt+L` goes back |
| `Super+G` | Group windows into tabs |
| `Super+S` | Scratchpad (`Super+Alt+S` sends a window there) |
| `Super+W` or `Super+Q` | Close window |
| `Super+Ctrl+V` | Clipboard history |

## 10 tips for coding with agents on Omarchy

1. **Give each job a workspace.** Workspace 1 for agents, 2 for your editor, 3 for the browser and docs. When an agent starts a long run, park it with `Super+Shift+Alt+1` and keep working.
2. **Use the built-in tmux layouts.** `tdl cx` opens Neovim, Claude Code and a shell side by side. `tdl c cx` adds opencode as a second agent (`icx` is the shortcut). `tds` gives you editor, a live diff watcher, shell and opencode in one window, and `tdlm cx` opens a `tdl` window for every subfolder of the current directory, one project per tmux window. Prefer Herdr, Omarchy's agent-aware multiplexer, over tmux? `hdl`, `hds`, `hdlm` and `hsl` build the same layouts there.
3. **One worktree per agent.** `ga <branch>` creates a git worktree and branch and drops you into it; `gd` removes it. Pair it with `tsl 4 cx` (four Claude Code panes in a grid) and every agent gets its own copy of the repo instead of fighting over one.
4. **Keep a scratchpad agent.** Send an agent terminal to the scratchpad with `Super+Alt+S`, then pop it over whatever you're doing with `Super+S` for quick questions.
5. **Screenshots straight into the agent.** Print Screen (or `Super+Ctrl+Alt+C` on keyboards without the key) copies a region to the clipboard; paste it into Claude Code or Codex with `Ctrl+V` (the agents read images from `Ctrl+V`, text from `Super+V`). The screenshot editor can redact before you paste.
6. **Turn error dialogs into text.** `Super+Ctrl+PrtSc` OCRs a screen region, which beats retyping a stack trace from a popup.
7. **Talk to it.** Pick a backend under *Setup > Defaults > Dictation* (Superwhisper is the default on fresh installs; Voxtype and Hex are the alternatives). Hold `Right Alt` to dictate a prompt, or `Alt+Space` to toggle. Installs from before 4.5 keep `F9` as push-to-talk.
8. **Get pinged when the agent needs you.** Claude Code supports a `Notification` hook; point it at `notify-send` in `~/.claude/settings.json` and Omarchy's notification center will tell you when it's waiting. Codex has a `notify` setting in `~/.codex/config.toml` that fires when a turn finishes.
9. **Keep keys out of your dotfiles.** Put API keys as session environment files in `~/.config/uwsm/env.d/`, which sits outside `~/.config/hypr`, so you can publish your Hyprland config without leaking anything. Better still, sign in to the agents with your subscription (OAuth) instead of exporting API keys at all.
10. **Let the agent rice your desktop, carefully.** Asking an agent to change your Hyprland config works well with the bundled Omarchy skill. If it goes sideways, `omarchy reinstall configs` puts the defaults back.

## Gotchas

- **Docker needs sudo by default.** An agent that runs `docker` will stall on a password prompt. Sudoless Docker is an opt-in under *Setup > Security > Sudoless Docker*, with the usual security trade-off (the docker group is effectively passwordless root).
- **Nvidia.** The installer chooses the right driver for your GPU generation and enables early KMS. On hybrid laptops where the Nvidia card doesn't drive the display, Omarchy deliberately skips some Nvidia video variables, so don't copy those into your config from old forum posts.
- **Screen sharing.** Native Wayland apps share fine through PipeWire. Apps running under XWayland (Discord is the usual one) can only see other XWayland windows; the Hyprland wiki's fix is `xwaylandvideobridge`.
- **Password managers in web apps.** Omarchy's web-app wrappers don't play well with 1Password, so log in through the regular browser first.
- **Installing on top of an existing Arch box.** Version 4 is designed to be installed from the ISO. The old curl-script install for vanilla Arch is documented for Omarchy 3; we couldn't confirm it's supported for 4.
- **Macs.** 4.5 adds native installs on M1, M2, M3 and Neo MacBooks through a Mac installer app or `brew install omacom/mac/omarchy`. The manual's Mac chapter still only covers Intel Macs, so expect rough edges.

## What changed in Omarchy 4.5 (Nov 5, 2026)

4.5 is the first big feature release since Quattro: 740-plus pull requests over about twelve weeks. *Update > Omarchy* (or `omarchy update`) on a 4.0.x install brings it in. The agent-relevant changes, from the release notes:

- **Several subscriptions per agent**, described above. Grok gets the same treatment as Claude and Codex.
- **Kimi and Ori join the lazy launchers**, Antigravity (`agy`) replaces Gemini, and the usage panel covers 23 agents and providers instead of 6.
- **Hotkeys move.** `Super+Grave` opens the Quake console, `Super+Q` also closes a window, `Super+L` cycles layouts, and inside a terminal `Super+C` / `Super+V` send `Ctrl+Shift+C` / `Ctrl+Shift+V`. The calculator moves to `Super+Shift+Q`, the activity monitor to `Super+Shift+T`, the network panel to `Super+Ctrl+N`, and the screenshot fallback is `Super+Ctrl+Alt+C`. Holding `Right Alt` dictates. Check your own bindings for collisions before upgrading.
- **The boot menu waits one second** instead of about six; keep tapping a key (down arrow works) to open it. That matters because the rollback snapshot lives there. Set `timeout: 5` in `/boot/limine.conf` if you want the old pause; updates leave it alone.

## From the terminal

Omarchy hands you a one-key, auto-approving agent on a fresh install. We respect the commitment. If that's your setup, you've earned the [--dangerously-skip-permissions](/d/skip-permissions/) tee. Keep backups anyway.

Sources: [omarchy.org](https://omarchy.org) · [Omarchy 4.5 release notes](https://omarchy.org/news/2026/11/omarchy-4-5/) · [Omarchy manual](https://omarchy.org/manual/) · [Omarchy source and releases](https://github.com/omacom/omarchy) · [Claude Code setup](https://code.claude.com/docs/en/setup) · [Claude Code hooks](https://code.claude.com/docs/en/hooks-guide) · [Codex CLI](https://developers.openai.com/codex/cli) · [Arch: openai-codex](https://archlinux.org/packages/extra/x86_64/openai-codex/) · [Google: Gemini CLI to Antigravity CLI](https://developers.googleblog.com/an-important-update-transitioning-gemini-cli-to-antigravity-cli) · [Arch: gemini-cli](https://archlinux.org/packages/extra/x86_64/gemini-cli/) · [Arch: opencode](https://archlinux.org/packages/extra/x86_64/opencode/) · [Arch: ollama](https://archlinux.org/packages/extra/x86_64/ollama/) · [Cursor CLI install](https://cursor.com/docs/cli/installation) · [VS Code on Linux](https://code.visualstudio.com/docs/setup/linux)
