Repolode
Submit
generalaction

generalaction/emdash

Desktop app for running multiple AI coding agents in parallel, each isolated in its own Git worktree and branch.

View on GitHubWebsite
Stars
6,029
Forks
638
Open issues
94
Last commit
13 hours ago

What is it?

Emdash is a desktop application for running AI coding agents side by side. Each task gets its own Git worktree and branch, so several fixes or features can be explored at the same time without agents colliding in one working copy. You review each agent's diff, then keep what works and merge it.

Emdash detects CLI agents already installed on your machine, including Claude Code, Codex, OpenCode, Amp, Cursor, and GitHub Copilot. It works with local repositories and with remote codebases over SSH/SFTP, using SSH agent, key, or password authentication with credentials kept in the OS keychain.

Issues from Linear, GitHub, Jira, GitLab, Asana, and several other trackers can be sent directly to an agent. From the same interface you can create pull requests, check CI status, and merge. App state lives in a local SQLite database, and telemetry is optional.

Key features

  • 01Runs multiple coding agents in parallel, each in an isolated Git worktree and branch.
  • 02Detects installed CLI agents such as Claude Code, Codex, OpenCode, and Amp.
  • 03Supports remote projects over SSH/SFTP with agent, key, or password authentication.
  • 04Imports tickets from Linear, GitHub, Jira, GitLab, and other trackers into agents.
  • 05Lets you review diffs, open pull requests, check CI, and merge in one place.
  • 06Stores app state locally in SQLite and makes telemetry optional.

Use cases

  • Trying several approaches to the same bug and comparing the resulting diffs.
  • Handing well-defined tickets from Linear or Jira to agents while you work on something else.
  • Running coding agents against code on a remote development server over SSH.
  • Keeping agent experiments isolated from your main working tree.

Quick start

  1. 1

    Install on macOS with Homebrew

    brew install --cask emdash
  2. 2

    Or download the desktop app for your platform from the releases page

  3. 3

    Launch Emdash and point it at a local project or remote machine; it detects installed agent CLIs automatically

Things to know

  • Requires at least one supported agent CLI (such as Claude Code or Codex) installed separately.
  • Agent CLIs may send code and prompts to their own providers, depending on the provider you choose.

Topics

#ai-agents#coding-agents#git-worktree#parallel-execution#desktop-app#ssh#llm#developer-tools#agenticdevelopment#agenticdevelopmentenvironment

FAQ

What is emdash?
Desktop app for running multiple AI coding agents in parallel, each isolated in its own Git worktree and branch. Emdash is a desktop application for running AI coding agents side by side. Each task gets its own Git worktree and branch, so several fixes or features can be explored at the same time without agents colliding in one working copy. You review each agent's diff, then keep what works and merge it.
Is emdash free to use? What license does it have?
emdash is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is emdash still maintained?
Actively maintained. The last commit to emdash was on Oct 19, 2026.
What can I use emdash for?
Trying several approaches to the same bug and comparing the resulting diffs. Handing well-defined tickets from Linear or Jira to agents while you work on something else. Running coding agents against code on a remote development server over SSH. Keeping agent experiments isolated from your main working tree.
How do I install emdash?
Install on macOS with Homebrew: brew install --cask emdash Or download the desktop app for your platform from the releases page Launch Emdash and point it at a local project or remote machine; it detects installed agent CLIs automatically
Where can I get emdash?
The source code and installation instructions are on GitHub: https://github.com/generalaction/emdash

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects