unstablebuild/rune
Developer ToolsA GPU-accelerated IDE and terminal multiplexer with a peer network for remote workspaces and coding agents.
What it is
Rune is a native IDE that doubles as a terminal multiplexer. It targets developers who want to move between projects and machines without losing their terminal-and-editor layout. On Linux it renders through OpenGL and on macOS through Metal, with no Electron shell.
Its organizing model is a screen multiplexer: nine workspace slots, plus unlimited terminals, tabs and windows. Instead of just a local editor, Rune instances can form an end-to-end encrypted peer network via headscale networking, so a laptop can open projects running on a workstation or headless node. Language intelligence packages are installed on demand, and the UI can host a built-in coding agent or work alongside external agents, which is why the project markets itself for both automatic and manual programming.
Quick start
On macOS or Linux, install the latest self-updating release:
curl -fsSL https://rune.build/install.sh | sh
Package-manager alternatives are Homebrew (brew install unstablebuild/tap/rune), a Debian/Ubuntu .deb from the latest release, or the rune-ide AUR package for Arch/Omarchy. Check prerequisites and first steps in the getting started guide.
To run a remote workspace, start a headless node in Docker:
docker run -d --name rune \
--hostname devpod \
-v /src/projects:/home/rune/projects \
unstablebuild/rune
docker logs -f rune # prints the sign-in code
Then, from a connected machine, open the mounted project:
workspaceopen rune://devpod/home/rune/projects
Development
Clone the repository, then run go run ./cmd/rune. make rune-dmg produces a macOS app bundle, and make -j16 compiles all executables into ./bin. Build requirements are in the building guide.
Constraints
- Desktop support is macOS and Linux; the install script and package options above assume those platforms.
- Installing with the script gets a self-updating release, while package-manager installs update through the package manager instead.
- The remote-workspace features require the Rune peer network and a configured node; headless deployment guidance covers service managers such as systemd, launchd, OpenRC and runit.
- Extensions can be installed from a Git repository or the official package index, so trust of community packages is up to the user.
- The project is GPL-3.0-or-later, which matters if you plan to modify or redistribute it.
HotRepo's take
Rune appeared at #10 on the Go daily list on 11 November 2026 with 38 stars added that day, against a total of 1,739 stars and 116 forks. That single snapshot shows it reaching the Go chart's top ten while still young, but gives no basis for claiming a longer trend. What makes it worth watching is the combination rather than one feature: a Go/TUI editor and terminal multiplexer with a native graphics backend and an encrypted machine network, aimed squarely at the remote-agent workflow that developers are already assembling from separate tmux, SSH and IDE tools.
Written by AI from the project's README, updated . The project's own docs are the source of truth.
You might also like
- herdrdev/herdrA terminal multiplexer that keeps multiple coding agents running across machines and tells you which one is stuck.
- eneskirca/nodetermOrganizes terminals and parallel AI coding agent sessions as persistent, draggable nodes on an infinite canvas, so you can see everything at once.
- ParadevAI/ParadevA desktop IDE that runs many CLI coding agents side by side in separate Git worktrees, with diffs and previews.
- anomalyco/runeA GPU-rendered Rust UI framework with its own renderer, text, layout, springs and headless rendering for macOS, Windows, Linux and web.
- eugenioenko/tttA terminal code editor for developers who want IDE-grade editing (LSP, multi-cursor, git) without leaving the shell.