enesalatas

I’m working on Spool, an open-source tool for running persistent Claude Code loops.

https://github.com/enes-alatas/spool

I wanted a way to manage multiple loops without babysitting terminal sessions and to talk to them as they work. With Spool, you can create and manage loops from a web UI, chat with them through Telegram or Slack, and let them communicate with each other. Each loop can run on your environment directly having access to everything you have, or in its own container or even better on Anthropic's cloud.

I’m using Spool to build Spool and figuring out what works well when several loops collaborate on the same project. That's also an interesting experiment going on.

I’d also love to hear how others are managing long-running or persistent agents.

1) Claude Max 5x, ChatGPT Plus (for minimal usage of Codex, mostly fired as a Pre/Post PR hook for Claude Code to get reviews from GPT models locally

2) None for now.

3) Claude Code CLI, but not always in a terminal I'm sitting in. I talk to it on Telegram and Slack for different projects.

4) Something custom that grew into a project: Spool (https://github.com/enes-alatas/spool). It runs long-lived Claude Code loops, each with a name and a mission. Each one wakes on its own timer or when someone messages it, works, and sleeps, and the session resumes with its context. Each loop gets its own container with an egress allowlist, and they message each other and me in a Telegram or Slack group. A web control room shows what every loop is doing and what it costs.

5) No, Spool covers it for me.

Thanks for stating the obvious but please see my comment too.

I've just seen your post, and I kind of went through similar pain, but went the other way round: a small self-hosted hub owns the Claude Code sessions, and the client is just Telegram, Slack or a web page, so a reconnect never touches a session. It does need a claude.ai login, though, same as Remote Control. Open source, in case you think it would help: https://github.com/enes-alatas/spool. I'd love to hear some feedback as well.

I shared Spool here a little while ago. It lets you create and manage Claude Code loops that you can talk to and have them collaborate over Telegram or Slack.

Now you can run those loops on Anthropic’s cloud without using your local compute resources. You can still talk to them and have them collaborate with loops running on your own machine.

Really glad you're enjoying it! Yeah, isolating loops from the local workspace, secrets, and MCPs was one of the things that needed to be done right from the beginning. It probably added a bit of complexity, but giving autonomous loops unrestricted access to everything just didn't feel right.

That’s actually an interesting idea. So far, I designed to use a single subscription since that was the case for me. But, just like you, teams might also want to have loops running on multiple subscriptions and collaborating. Let me think what the simplest could be to do this without making the onboarding too complicated. If you could spare time, that would be also great if you could file an issue on this to get spool’s own loops involved.

I built Spool because I wanted persistent Claude Code loops I could talk to, without keeping track of separate terminal sessions.

It makes it easier to create and manage loops, and lets them communicate with you and each other.

I'd love to hear whether you'd find this useful.