If you're more of a SWE than a data scientist, check out Mjolnir: https://github.com/BrokkAi/mjolnir/

Built in containerization and EC2 support, allows moving sessions across harnesses/profiles/machines

Goodness - triggers my ADHD. Power to those who find it useful, I would be so lost in the wrong details (the quota screen is pretty cool).

Same. That many panes is fine until something breaks at 3am and you're hunting for the one that matters.

When all you have is a drawer-full of hammers, get yourself a mythical hammer. Clever name.

It's a 2 day old project. https://github.com/ZhengyiLuo/AgentsDock

That is "from last year" in human time.

Matches my experience. Agent-built side projects of mine go from empty repo to stale docs in a week, easily.

Created a new repo to open-source since the old one has too many commits and some uncessary histories haha

That is new repo for open source, we've actually worked on it for a while

I went from IDE (VS Code) to terminal (wezterm) running half a dozen agents in an app called Herdr. My computer work is mostly in the terminal now, prompting agents. And they have access to my Obsidian vaults. Can read, and add to them. And access all my notes via a local RAG. I neither need projects in Claude, nor IDEs. Just agents accessing my local vaults and code repos.

Yes, we need agent-oriented IDE not the old VS code style!

I disagree. Reviewing six agents' diffs in a terminal gets miserable fast. You still need an editor, just built for reading code.

Isn’t this just paseo.sh with less features and tightly-coupled tailscale?

I dont care about codex,cc harness. but I do miss a proper terminal in pi web; I will likely endup forking it... the lamma cpp built-in web app is super nice, but also needs to be hacked to support api providers, and also no terminal :/ .

I think the definitive solution way would be using vscode stripped for web (what gh codespaces use) and add pi to it.

Looks useful! curious how you would compare it against something like Paseo?

Tried it out, missing a few important features we need : ). Paseo is great though

Does either one handle resuming a crashed agent run mid-experiment? That's where I'd expect the gap to show, not in the UI.

IDEs are today's JS frameworks

Looking at the screenshots, this looks close to what an agent factory for teams could look. Agents on multiple servers, people can see each conversation. Test it…

Let me know how it goes!

- arent there 10000 projects that do this

- what makes this one special?

Geared toward for AI research so it is has a lot of nice-to-have features.

When you move the app to the trash, it leaves the server you have to install. It's still running and set to restart automatically.

I suggest putting the tests in a separate folder ;)

hhh you spotted my messy folders ;)

Multi-agent tools need more than a visual dashboard

looks interesting! Trying it out. Thanks

agentslop

Trying this now

Thanks!! Let me know if you need any help

Every comment here recommends a different agent tool. Agentic research is mostly just shopping.