Claude Cowork Is Basically a Safe(er) Version of Open Claw.

If you’ve been following the OpenClaw hype — AI that wakes up and does things on its own, on a schedule, without you asking — that’s what has everyone excited. Scheduled tasks. Not “ask AI to do a thing.” AI that does things while you sleep. That’s a fundamentally different relationship with a tool.

I trust Cowork because Claude Code has been super reliable for me. Hundreds of sessions, thousands of messages, 285 files touched. It hasn’t been destructive. So when Cowork launched with scheduled tasks built in, I decided to set it up as a daily productivity helper (you also need to enable Anthropic’s productivity skill which introduces an OpenClaw-like memory system, a big part of making it more useful as a personal assistant).

I’ll confess: I didn’t actually know what I wanted it to do for me. But the onboarding was pretty slick and I quickly landed on a number of things — daily briefings, calendar scanning, email flagging, tracking whether I’m keeping up with LinkedIn. What surprised me was that even though I had a notional idea of what I wanted, Claude helped me see there are things it can do to take the load off that I hadn’t thought of.

Like school emails. I have three kids. Their schools send roughly one million emails per week, and my system for processing them has been “skim subject line, tell myself I’ll read it later, forget it exists.” Cowork flagged a permission slip due in three weeks that I absolutely would have missed. My wife is super organized and methodical, so basically I’d just depend on her to catch it — and hear about it later.

I definitely needed my technical expertise to work around a few limitations. This isn’t plug-and-play for everyone yet. I’ll report back in a few weeks and let you know how it goes. But I’m looking forward to getting my first daily briefing tomorrow morning.