You may remember us from our early days, when we made a remote-control phone app for Claude Code and Codex. Since the harness ran on a laptop back then, we had to build a relay to reach it, it stopped when the laptop closed, and moving it to a cloud sandbox meant syncing code and agent state between machines. We didn't think remote-controlling a local agent was end-game for how people would run agents, so we built a new product where the agent runs in the cloud, any client can talk to it through an API, and it can still use your laptop as one of its machines.

Anthropic and OpenAI are building hosted versions of this, like Claude Managed Agents and the OpenAI Agents API. Omnara is an open-source version you can run on your own servers and use your own models. It's basically a harness server like the opencode or codex servers, but durable, scalable, with the usual harness features like file and shell tools, subagents, compaction, permission prompts, MCP, skills, memory, yada yada yada.

Here's some interesting architectural decisions we made while building this out:

Most harnesses, like Claude Code, Codex, OpenCode, Pi, Goose, etc, run the agent loop in one long-lived process that keeps its log in a JSONL or SQLite file. In Omnara, the log is an append-only event log in Postgres, with inputs, model outputs and tool calls stored as typed rows in their own tables. The loop runs on stateless workers in the control plane. A worker runs one iteration of the loop, either a model call or the tool calls it produced, then releases the agent and puts the next iteration on a queue in Postgres. Whichever worker is free picks it up, so each iteration of an agent's loop can run in a different process on a different machine. Each iteration's result commits in one transaction, so the model output, its tool calls, usage and the next wakeup are saved together or not at all. If a worker dies mid-iteration, nothing commits, its lease expires, and another worker retries it. Agents waiting on a user, a tool or a retry don't hold a worker, so one worker can multiplex thousands of agents.

Also unlike other harnesses, the machines that the workers (our agent loop) run on aren't where the agent runs its commands. Instead, any macOS or Linux machine with an internet connection, like your laptop or any cloud sandbox, can run a small daemon (omnarad) that connects out to Omnara, so it doesn't need any open inbound ports. The agent then gets tools for that machine, like running a command or editing a file. An agent can have any number of machines connected to it. A fun thing I like to do is I have my laptop and a cloud sandbox connected to the same agent, so when my laptop is on, my agent codes there and I can interact with it, and I have it often push to Github. When I close my laptop, my agent sees the laptop went offline, pulls the latest changes to its cloud sandbox, and continues there.

The entire codebase is open source at https://github.com/omnara-ai/omnara, and it’s self hostable, even on a single machine with docker compose. You can also use the hosted version at https://omnara.com. Let me know what you think!

Comments URL: https://news.ycombinator.com/item?id=50183892

Points: 57

# Comments: 15

https://github.com/omnara-ai/omnara - this is a self hostable agent API that I'm working on. It stores the state of all agents in a postgres db you can easily query, rather than a local json file or sqlite file per agent.

Comments URL: https://news.ycombinator.com/item?id=49524438

Points: 5

# Comments: 2

Now we use the Claude Agent SDK (basically a headless version of Claude Code), and we make our own UI for laptop and mobile. This is way easier to maintain than the previous solution we had. You can import Claude Code CLI sessions into Omnara, but you cant see a 1:1 realtime Claude Code CLI session in Omnara anymore. But we think that a GUI is a better experience than the CLI anyways for managing a bunch of agents.

Is it possible to completely disable or not use the remote sandbox features? I would never use them and would prefer my code stays on my device.

Yes, the remote sandbox feature is disabled by default, and you have to manually enable it for the syncing to start.

For those of us that are using subscriptions, does it show our remaining usage? I would hate to run out of tokens in the middle of a session.

Currently Omnara doesn't show your usage limits, you would have to check that at claude.ai. I'll look into add that though.

One feature of the CC TUI I sorely missed on mobile is the ability to look up and directly reference files via “@“. Is any functionality like this planned?

Yes, this exists in Omnara already!

Are there any plans to offer a cheaper tier for those of us that just want to run this on our own machines?

That's a good idea, we'll think about doing this where we don't offer sandbox + voice, and just have the messaging service.

What do you mean by syncing? Happy coder syncs sessions between all my happy coder clients. I can even see in real time how happy coder in my browser's conversations progress as well as on my phone, in parallel.

Omnara also displays realtime conversations between all Omnara clients. What I mean by syncing is syncing your conversation and code changes to a cloud sandbox, which is useful if you're using Omnara on your laptop and you close your computer (as explained in the original post). If you run your agents on a persistent cloud VM, then this is less of a value add.

I can voice chat with Happy coder.

We use https://docs.livekit.io/agents/ which runs the voice agent in the cloud (to enable the above use case, and a better experience when you're using your phone when it's off), whereas I believe happy runs a client-side voice agent.