Emdash

Developers who use AI coding agents like Claude Code, Codex, or Cursor often juggle them one at a time, switching tools, terminals, and contexts for each task. There is no shared place to run several agents in parallel, track their progress across isolated branches, preview the resulting app, or review diffs and CI checks before shipping. Issues from Linear, Jira, GitHub, Notion, or Asana have to be manually copied into an agent's context. Teams also lack a simple way to schedule recurring agent work (like bug sweeps or release prep) or to run agents on remote machines with proper workspace isolation.

Emdash is a free, open-source agentic development environment that runs coding agents in parallel, each inside its own Git worktree with its own branch, terminal, and conversation history. It pulls issues directly from tools like Linear, Jira, GitHub, GitLab, Notion, and Asana so agents get full context, includes a built-in browser to preview local or remote apps, and provides a review flow for diffs, PRs, and CI checks without switching tools. It supports scheduling recurring automations, connecting to remote machines via SSH with full worktree isolation, and provisioning on-demand workspaces through setup/teardown scripts. A built-in library lets users save prompts, install skills, and sync MCP servers across 25+ supported coding agents including Codex, Claude Code, Amp, Antigravity, Cursor, Droid, Hermes, and Pi.