Agents In Use
What’s new todayLast update
108 use cases added today

Claude to build a border trip logger for Android

Michael de Haas built his first Android application with Claude Code in less than twenty four hours. WorkTravelLogger was intended to record border trips and present a rolling day counter, rather than asking him to reconstruct that history at the end of each month.

Source published linkedin.comAgent: Claude
OUR ADAPTATION

Suggested prompt

Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.

Build an Android trip log with local storage, geofenced border events and monthly CSV or PDF exports. Present each phase for my review and explain any build failures.

Evidence at a glance

Sourced account
Source author
Michael de Haas
Original account
Read the source ↗
Published
Source link checked
Reported result
He reported a functioning Android trip logger built within a day, with monthly PDF and CSV exports; real crossing records were not published.
Human involvement
De Haas reviewed every phase and checked fixes for version conflicts, package structure and emulator problems.

A linked account supports attribution. It does not mean we independently reproduced its result.

Report an issue with this caseOpen correction email

Opens your email app with the case link included. Nothing is sent automatically.

How the agent did it

  1. Define the border event log, rolling counter and monthly export requirements.
  2. Review the architecture and each implementation phase with Claude Code.
  3. Resolve version, package and emulator problems before testing the Android application.

Michael de Haas built his first Android application with Claude Code in less than twenty four hours. WorkTravelLogger was intended to record border trips and present a rolling day counter, rather than asking him to reconstruct that history at the end of each month.

He described a phased build with a reviewed architecture, local storage through Room and SQLite, and four Jetpack Compose screens. The reported application included geofenced border events, a rolling six month counter and monthly PDF or CSV exports through Android's share sheet.

The work was not a single prompt followed by an untouched result. De Haas reviewed every phase and worked through implementation problems with the agent. Those included a KSP version conflict, a mismatch between package names and directory structure, and emulator issues. His account emphasizes methodical correction and checking the application as the build progressed.

The outcome was a functioning application as reported by its author. The post does not provide a log of real border crossings or an immigration authority's assessment of its calculations. It should therefore be read as a personal travel record tool, not validated proof of legal status. The concrete output is local trip recording with exportable monthly records, produced through an iterative build under human review.

Prompt
Suggested prompt (written by Agents In Use, not the author's)

Source: www.linkedin.com · Michael de Haas

Frequently asked questions

Where did the application store its records?

The described architecture used Room and SQLite for local storage.

What required human review during the build?

Every phase was reviewed, including fixes for KSP versions, package directories and emulator problems.

Does the post prove legal compliance?

No actual crossing history or official assessment appears in the account.

More details

About this account

Claude Code generated and revised the Android application, including local records, geofences and export features.

Firsthand source opened and read. Suggested prompt written by us from the reported workflow; no original task prompt is quoted.

Credit: Michael de Haas

Source published

Added to Agents In Use

Source checked

Tools and topics

Claude Code · Kotlin · Jetpack Compose · Room · SQLite

Border trip log · Android geofences · CSV export · Monthly PDF

Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.