Until last week, every AI task was reactive. You asked, it answered. Last week, that changed quietly — and it's going to change how work gets done in your company.
Anthropic shipped scheduled tasks for Claude: the ability for an agent to run on its own schedule, without anyone prompting it. Every Friday at 5pm, pull this report. Every time a new ticket comes in, triage it. Every morning at 7am, scan the pipeline and brief the leadership team.
That sounds like a small feature. It's not. It's the line between an AI that's a tool and an AI that's a teammate.
Why schedules change everything
A synchronous AI — the one you talk to — is useful but limited. It waits for you. A scheduled agent works while you sleep. The difference in throughput isn't linear. It's categorical.
A synchronous agent is a smart intern who needs to be told what to do. A scheduled agent is a teammate who already knows.
Three workflows this unlocks today
-
01
Morning briefings, compiled before you're awake.
The agent scans your inbox, pipeline, calendar, and Slack overnight. By 7am you have a one-page summary on your phone.
-
02
Recurring reports that write themselves.
Weekly sales recap. Monthly board update. Quarterly client review. All pulled from source data, drafted in your voice, waiting in your inbox.
-
03
Proactive alerts on the stuff that matters.
Agent watches your key accounts, flags usage drops, pipeline stalls, or renewal risks before anyone notices.
What to do with this
If your team runs on "Friday reports" or "Monday meetings" or "end-of-quarter summaries," those workflows are now automatable. They were always automatable in theory. Now they're automatable in practice — cheaply, reliably, and without asking a developer.
Book a 30-minute strategy call
Tell us one recurring task your team does every week. We'll sketch what it looks like as a scheduled agent.
Schedule the call →