Emdash is a free, open-source agentic development environment backed by YC W26, available for macOS, Windows, and Linux. Its core purpose is to let developers run and manage coding agents, such as Claude Code and Codex, within a unified platform, and it provides capabilities such as task isolation, scheduling, review, and remote execution around agentic workflows.
Main capabilities
- Run agents in parallel: Multiple coding agents can run at the same time, without being limited to a specific agent, avoiding lock-in.
- Task and context integration: Issues can be passed directly to agents, with support for pulling tasks from platforms such as Linear, Jira, GitHub, Notion, and Asana so agents get full context.
- Isolated workspaces: Parallel tasks run in independent Git worktrees, and each task keeps its own branch, terminal, conversation, and review state.
- Built-in app preview: Local apps, remote previews, and browser profiles can be opened within a task.
- Change review: Diffs, PRs, and checks can be inspected, and users can commit, push, open pull requests, and monitor CI without switching tools.
- Scheduled scheduling: Recurring agent runs can be scheduled to automate repetitive work such as bug scanning, release preparation, review, and checks.
- Remote execution: Agents can run remotely by connecting via SSH to development servers, cloud VMs, or GPU machines, while maintaining full worktree isolation.
- On-demand workspace configuration: Setup and teardown scripts can create isolated local, remote, or cloud environments for each task.
- Prompt, skill, and MCP tool management: Prompts, skills, and MCP tools can be managed centrally in one place.
How to use Emdash is used as a development environment and supports installation on macOS, Windows, and Linux. It allows the use of any coding agent; details such as the specific list of supported agents and whether API keys are required are not fully explained in the provided materials.
Use cases Suitable for developers and teams that need to handle multiple coding tasks at the same time, want to hand issues directly to agents, need to run tasks in isolated environments, need to execute agents remotely or in the cloud, and need to centrally review codechanges and automate repetitive development work.
Limitations The materials do not provide specific installation commands, the complete list of supported agents, API key requirements, how code privacy is handled, pricing (it only states that it is free and open source), or a clear explanation of whether it is related to the punctuation mark em dash; this information is not currently provided.
