alt Hacker NewsShow HN: Emdash – Open-source agentic development environment
Hey HN! We’re Arne and Raban, the founders of Emdash (https://github.com/generalaction/emdash).
Emdash is an open-source and provider-agnostic desktop app that lets you run multiple coding agents in parallel, each isolated in its own git worktree, either locally or over SSH on a remote machine. We call it an Agentic Development Environment (ADE).
You can see a 1 minute demo here: https://youtu.be/X31nK-zlzKo
We are building Emdash for ourselves. While working on a cap-table management application (think Stripe Atlas + Pulley), we found our development workflow to be messy: lots of terminals, lots of branches, and too much time spent waiting on Codex.
Emdash puts the terminal at the center and makes it easy to run multiple agents at once. Each agent runs as a task in its own git worktree. You can start one or a few agents on the same problem, test, and review.
Emdash works over SSH so you can run agents where your code lives and keep the parallel workflow. You can assign tickets to agents, edit files manually, and review changes.
We also spent time making task startup fast. Each task can be created in a worktree, and creating worktrees on demand was taking 5s+ in some cases. We now keep a small reserve of worktrees in the background and let a new task claim one instantly. That brought task start time down to ~500–1000ms depending on the provider. We also spawn the shell directly and avoid loading the shell environments on startup.
We believe using the providers’ native CLIs is the right approach. It gives you the full capabilities of each agent, always. If a provider starts supporting plan mode, we don't have to add that first.
We support 21 coding agent CLIs today, including Claude Code, Codex, Gemini, Droid, Amp, Codebuff, and more. We auto-detect what you have installed and we’re provider-agnostic by design. If there’s a provider you want that we don’t support yet, we can add it. We believe that in the future, some agents will be better suited for task X and others for task Y. Codex, Claude Code, and Gemini all have fans. We want to be agnostic and enable individuals and teams to freely switch between them.
Beyond orchestration, we try to pull most of the development loop into Emdash. You can review diffs, commit, open PRs, see CI/CD checks, and merge directly from Emdash once checks pass. When starting a task, you can pass issues from Linear, GitHub, and Jira to an agent. We also support convenience variables and lifecycle scripts so it’s easy to allocate ports and test changes.
Emdash is fully open-source and MIT-licensed.
Download for macOS, Linux or Windows (as of yesterday !), or install via Homebrew: brew install --cask emdash.
We’d love your feedback. How does your coding agent development setup look like, especially when working with multiple agents? We would want to learn more about it. Check out our repository here: https://github.com/generalaction/emdash
We’ll be around in the comments — thanks!
Comments
Here's my question:
if agents continue to get better with RL, what is future proof about this environment or UI?
I think we all know that managing 5-10 agents ... is not pretty. Are we really landing good PRs with 100% cognitive focus from 5-10 agents? Chances are, I'm making mistakes (and I assume other humans are too)? Why not 1 agent managing 5-10 agents for you? And so on?
Most of the development loop is in bash ... so as long as agents get better at using bash (amongst other things), what happens to this in 6 months?
I don't think this is operating at a higher-level of abstraction if agents themselves can coordinate agents across worktrees, etc.
So, what's your business model ? Is this an YC product, or a tool you developed while working on a YC product ?
Been driving my agents (CC, currently testing Pi) for a couple of weeks via Emdash. Finally, got a productive worktree setup working. There were still rough edges when I started, but the team has shipping fast [0] and is vaporizing concerns on the fly. Building on top of the native CLI seems to be the right strategy as well.
the worktree pre-warming detail is interesting -- keeping a reserve pool and letting new tasks claim one instantly is the same pattern as connection pool pre-warming in databases. the underlying bottleneck is probably git having to traverse pack files and update the index when you run 'git worktree add'. one thing worth trying if you haven't: sparse checkout on the worktrees can cut that initialization time further, especially in large monorepos where most files are irrelevant to a given agent task.
this looks great, but can't test, the .deb package is broken with an issue about NODE_MODULE_VERSION mismatches. There seems to be a PR waiting for approval. Will keep an eye on it.
Have you considered adding any kind of agent coordination layer, e.g. letting one “orchestrator” agent spawn and direct sub-agents on specific subtasks, rather than having the developer manually assign each task? Or is the explicit human-in-the-loop assignment a deliberate design choice to keep control and avoid runaway costs?
How does Emdash handle state management when running multiple agents on the same codebase? Particularly interested in how you prevent conflicts when agents are making concurrent modifications to dependencies or config files. Also, does it support custom agent wrappers, or do you require the native CLI?
Let's go! Love that this is a solid OSS alternative to what's already out there!
i'll have to give it a shot, the market needs an open source cursor right now
Looks cool! Thank you for sharing.
Congrats on the launch
LFG!
[dead]
Please codesign your Windows installer exes :)