Claude Code · Anthropic
Plan your day from your calendar and triage email in the terminal
A founder and writer built a personal assistant in Claude Code that plans each day around his calendar, sorts unread email and drafts replies for approval, and runs a weekly review.
What they did
The assistant reads today's calendar, finds open time and fits one to three priorities around existing commitments with some buffer, then writes a daily file. For email it sorts unread mail into reply, review and archive, pulls past conversation before drafting, and logs commitments as follow ups. Other commands produce meeting briefs, a next action suggestion, stale project flags and an end of day log.
How it went
The poster says the plans now fit real calendar constraints and stalled projects get surfaced, though he gives few measured results.
Worth knowing
Markdown storage caused duplicates and drift until he moved records to JSON, and the setup stops when the laptop is off. Sending email always needs his explicit approval.
Try it with Claude Code
Read my calendar for today from <your calendar> and my open tasks in <your task file>, then propose a plan with one to three priorities that fit the free time and leave buffer. Next, go through my unread email in <your inbox>, sort it into reply, review and archive, and draft replies for the reply pile using past threads for context. Do not send anything; list the drafts for my approval.
We wrote this prompt from the case. Replace the parts in angle brackets with your own details.
What this label means: User reports
Someone using it at work posted about it. Any numbers are theirs, so we write "the poster says".
Real cases of AI agents at work, from every major agent, each linked to where it was first reported. Rewritten in our own words, most with a prompt you can try. These are not Wize customers.