Hacker News new | ask | show | jobs
by Aurornis 124 days ago
> You basically got 95% of the way there with Claude Code inside of a container.

OpenClaw and Claude Code aren't solving the same problems. OpenClaw was about having a sandbox, connecting it to a messenger channel, and letting it run wild with tools you gave it.
2 comments
koolala 124 days ago
A messenger and ssh'ing into Claude Code from your phone aren't that much different.
link
Aurornis 124 days ago
I’m not an OpenClaw user but it’s obvious that OpenClaw was very different than that.

OpenClaw was about having the agent operate autonomously, including initiating its own actions and deciding what to do. Claude Code was about waiting for instructions and presenting results.

“Just SSH into Claude Code” is like the famous HN comment that didn’t understand why anyone was interested in DropBox because you could do backups with shell scripts.

link
koolala 123 days ago
You have used Claude though? Or codex or antigrav? backups seem different
link
theturtletalks 124 days ago
The real magic is heartbeat which is essentially cron on steroids. The real difference between running Claude Code in the terminal and OpenClaw is that the agent is actually intuitive and self-driven.

People would wake up to their agent having built something cool the night before or automate their workflow without even asking for it.

link
koolala 123 days ago
Is the idea it can't handle 24hr+ tasks? I know Codex can do heartbeat but I haven't tested the limits of its heart.
link
lvl155 124 days ago
That’s what CC does…I don’t need a messenger wrapper to do those things.
link