Claude to import school dates in a family's style
Zach Smith's family calendar held school events, practices and birthdays, but importing a new school year remained an annual manual chore. The school's PDF used small month grids and a color key. Smith normally interpreted each colored square, named the event, chose single or multi-day blocks and decided whether a specific time belonged.
linkedin.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Study last year's school events in our family calendar and explain our conventions. Then import this new school PDF using those conventions, asking about inconsistencies before creating the affected entries.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Zach Smith
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- Smith reports 19 styled events in five minutes and a further two-minute refinement extending three school breaks across adjacent weekends.
- Human involvement
- Smith answered clarifying questions and requested weekend extensions; no complete independent date audit is described.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Let Claude read the previous year's school events before giving it the new PDF.
- Have it explain the inferred naming, timing and multi-day conventions.
- Import the new events, answer inconsistencies and refine how breaks appear.
Zach Smith's family calendar held school events, practices and birthdays, but importing a new school year remained an annual manual chore. The school's PDF used small month grids and a color key. Smith normally interpreted each colored square, named the event, chose single or multi-day blocks and decided whether a specific time belonged.
Instead of prescribing a new naming convention, he first asked Claude Code to study the previous year's school entries. The agent had read, create and edit access to his calendars. It described his existing pattern back to him: capital letters for closures, title case for milestones, combined early-dismissal and last-day entries, and different handling of timed conferences.
After receiving the new PDF, Claude created 19 events in about five minutes. Smith then asked for Thanksgiving, winter and spring breaks to include adjoining weekends so they would be visible in any week view. That refinement took about two more minutes.
The source says Claude asked clarifying questions where earlier behavior was inconsistent. It does not describe a full independent audit of every imported date. The concrete output was a populated shared calendar that reused the family's own conventions, followed by an explicit change to how longer breaks were represented.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: www.linkedin.com · Zach Smith
Frequently asked questions
Why give last year's calendar before the new PDF?
Claude learned the family's existing naming, timed-event and multi-day conventions from actual entries.
What changed after the first import?
Thanksgiving, winter and spring breaks were extended across adjoining weekends so they appeared in any week view.
How were inconsistencies handled?
The agent asked clarifying questions where past behavior was inconsistent; the source does not report a full date audit.
More details
About this account
Claude inspected historical calendar events, inferred conventions, read the color-key PDF and created 19 entries.
Firsthand source opened and read. Suggested prompt written by us from the reported workflow; no original task prompt is quoted.
Credit: Zach Smith
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code · Shared calendar · School PDF
School calendar · Historical conventions · PDF event import
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.