Skip to content
MIYA·AI·LAB / generative-AI test logs LLM · Claude Code · MLX · Ollama
AI MiyaAILab_
  • // Lab
  • // all logs
JA / EN
← Miya-Gadget
Generative AI PC

Easy Task Management with Claude Code — Building a Mobile Budget App from Your Smartphone

2025年11月19日 ·
Easy Task Management with Claude Code — Building a Mobile Budget App from Your Smartphone

In the previous article, we built a command-line budget app. But didn’t you think “it would be nice to quickly log expenses from my phone”?

I often find myself wanting to check “how much have I spent so far?” while out and about. So this time, I built a mobile-optimized web budget app.

Since it runs in a browser, you can access it from a PC, smartphone, or tablet — anywhere. With Claude Code, building web apps is surprisingly easy!

 

Table of Contents

Toggle

  • Recap of the Previous Article
  • What We’re Building: A Mobile-Optimized Web Budget App
      • Core Features
      • Mobile Optimization Features
  • Creating the Project with Claude Code
      • Creating the Project Folder
      • Instructions for Claude Code
  • Installing Streamlit
      • Installation Command
  • Launching the Web App
      • How to Launch
      • Accessing from Your Smartphone
  • User Experience on Mobile
      • Recording an Expense
      • Totals and Category Breakdown
  • Customizing the Design
      • Customization Request
  • Deploying to Streamlit Cloud
      • Deployment Steps
  • Pitfalls and Solutions
      • Screen Width on Mobile
      • Data Persistence
  • When to Use CLI vs Web Version
      • When the Web Version Shines
      • When the CLI Version Shines
  • Summary
      • Key Takeaways

Recap of the Previous Article

the previous article, we built a command-line budget app.

While it had graph display functionality and was quite useful, it wasn’t practical for recording expenses on the go.

This time, we’ll solve that problem by building a web version you can use anytime, anywhere. It’s optimized for mobile screens, so touch input feels smooth and natural.

 

What We’re Building: A Mobile-Optimized Web Budget App

Here are the specifications for the web budget app we’re building.

Core Features

  • Expense Recording: Easily input category, amount, and notes
  • Monthly Expense List: Displayed in reverse chronological order
  • Total Display: See this month’s total spending at a glance
  • Category Breakdown: See where you’re spending the most

Mobile Optimization Features

  • Responsive Design: Automatically adjusts to mobile screens
  • Touch-Friendly: Large buttons for easy input
  • Simple UI: Shows only what you need
  • Instant Updates: Entries appear in the list immediately

Using the Streamlit framework, you can build modern web apps with Python alone.

 

Creating the Project with Claude Code

Let’s get started building with Claude Code right away.

Creating the Project Folder

  1. Create a folder called “web_budget” on your desktop
  2. Open PowerShell or Command Prompt
  3. Run the following commands
cd Desktop/web_budget
claude

Once Claude Code starts, give it these instructions:

Claude CodeへのWeb家計簿アプリ作成依頼

Instructions for Claude Code

Streamlitを使ってスマホに最適化したWeb家計簿アプリを作ってください。

機能:
1. 支出の記録(カテゴリ、金額、メモ)
2. 今月の支出一覧を表示
3. 今月の合計金額を表示
4. カテゴリ別の集計を表示

要件:
- スマホ画面に最適化されたレイアウト
- タッチ操作しやすい大きなボタン
- データはJSONファイルに保存
- シンプルで見やすいデザイン

Claude Codeがこの指示を受けて、Webアプリを作成してくれます。

私の環境では、`app.py`というファイルが作成され、Streamlitのインストール手順も提示されました。

 

Installing Streamlit

To run the web app, you need the Streamlit library.

Installation Command

pip install streamlit

Running this command automatically installs Streamlit and its dependencies.

Once installation is complete, you can launch the web app.

 

Launching the Web App

Let’s actually launch the web budget app we created.

How to Launch

streamlit run app.py

Running this command automatically opens your browser and displays the web app.

Web家計簿のメイン画面

URLは通常 `http://localhost:8501` です。このURLをスマホのブラウザで開けば、スマホからもアクセスできます。

Accessing from Your Smartphone

同じWi-Fiネットワーク内であれば、スマホからもアクセスできます。

  1. PowerShellに表示される「Network URL」を確認
  2. スマホのブラウザでそのURLを開く
  3. 外出先でも使いたい場合は、Deploying to Streamlit Cloud

 

User Experience on Mobile

Using it on an actual smartphone, you can see it’s well-optimized for touch input.

スマホ表示画面

Recording an Expense

Enter the category, amount, and notes using the form at the top of the screen.

Categories are selected via dropdown, making touch operation easy. After entering the amount and notes, tap the “Record” button.

Entered data appears instantly in the list below. This immediate feedback gives it a native app-like feel.

Totals and Category Breakdown

集計表示画面

The top of the screen shows this month’s total spending in large text.

Category breakdowns are also displayed, making it easy to spot trends like “I’ve been spending a lot on food this month.”

 

Customizing the Design

The basic functionality was complete, but I wanted to improve the look a bit.

So I asked Claude Code for additional customizations.

Customization Request

カラーテーマを追加してください。
また、支出の削除機能も追加したいです。

Claude Code responded quickly, adding colorful category color-coding and a delete button for each expense entry.

The color coding makes categories much easier to identify at a glance.

 

Deploying to Streamlit Cloud

Running it locally is convenient, but you’d want to use it on the go too, right?

With Streamlit Cloud, you can publish your web app for free.

Deployment Steps

  1. Create a GitHub repository
  2. Push the app you created
  3. Connect the repository in Streamlit Cloud
  4. Click the deploy button

Once deployment is complete, `https://あなたのアプリ名.streamlit.app` a URL like this will be issued.

Add this URL to your smartphone’s home screen and it works just like a native app.

 

Pitfalls and Solutions

There were a few stumbling blocks during development.

Screen Width on Mobile

Initially, horizontal scrolling occurred when viewing on mobile.

This was caused by Streamlit’s default width settings. When I told Claude Code “there’s horizontal scrolling on mobile,” it `st.set_page_config`added code to adjust the layout.

Data Persistence

Since Streamlit re-executes on every page reload, data storage requires careful consideration.

JSONファイルに保存することで、リロードしてもデータが残るようになりました。Deploying to Streamlit Cloudする場合は、データベースを使う方が良いかもしれません。

 

When to Use CLI vs Web Version

After building the web version, the use cases for each version became clearer.

When the Web Version Shines

  • Recording expenses on the go
  • Quick input from your smartphone
  • Sharing with family members

When the CLI Version Shines

  • Detailed graph analysis
  • Exporting data as CSV
  • Batch processing large amounts of data

Personally, I find it convenient to use the web version for daily recording and the CLI version for end-of-month analysis.

 

Summary

This time, we built a mobile-optimized web budget app using Claude Code.

Key Takeaways

  • Streamlit lets you build web apps with Python alone
  • Responsive design optimized for mobile
  • Large touch-friendly buttons and simple UI
  • Free hosting on Streamlit Cloud
  • Claude Code supports everything from library selection to implementation

Next time, I’d like to build a more practical tool — an automatic price monitoring tool for sites like Amazon.

Even programming beginners can easily build app-like web applications using Claude Code. Try customizing it for your own needs!

You might also like

More generative-AI logs from the lab.

  • Auto-Generate PDF Reports with Claude Code: From Raw Data to Polished Documents
  • Can you really cut your AI API bill? I deployed the context-compression tool “Headroom” and measured it
  • Automate Email Sending with Claude Code: Fully Automating Routine Tasks
  • Dual Tesla V100 SXM2 on a Single PCIe Slot: 64GB VRAM & 300 GB/s NVLink Tested — Is This $700 Setup Worth It?
Previous Article Build a Budget Tracker App with Claude Code: Data Visualization Made Easy
Next Article Smart Shopping with Claude Code: Building an Automatic Price Monitoring Tool

Related Posts

Building a Weather Forecast App with Claude Code and API Integration

Building a Weather Forecast App with Claude Code and API Integration

Auto-Generate PDF Reports with Claude Code: From Raw Data to Polished Documents

Auto-Generate PDF Reports with Claude Code: From Raw Data to Polished Documents

Automate File Organization with Claude Code: Tidy Up Messy Folders in Seconds

Automate File Organization with Claude Code: Tidy Up Messy Folders in Seconds

Building a Full-Stack Blog System with Claude Code: Applying Everything I Learned

Building a Full-Stack Blog System with Claude Code: Applying Everything I Learned

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Shichinomiya

Shichinomiya

A blogger who loves PC and gadgets. Sharing daily discoveries.

@shichinomiya_s

Popular Posts

  • Does the trending Claude Code skill “ADHD” actually make the agent smarter? A measured duel vs single-shot
  • Your First Program with Claude Code! Building a Game Through Conversation
  • Modded RTX 4080 32GB Benchmarked: Qwen3.8-27B at 262K Context, 125B MoE, and MiniMax H3 Video — What 32GB Actually Delivers
  • The Complete Guide to Claude Code: Get Started with AI Development on Windows in 5 Minutes
  • Dual Tesla V100 SXM2 on a Single PCIe Slot: 64GB VRAM & 300 GB/s NVLink Tested — Is This $700 Setup Worth It?

Categories

  • Announcements
  • Cars
  • Cycling
  • Gadgets
  • Generative AI
  • Home Appliances
  • Internet Service
  • Outings
  • Overseas Shopping
  • PC
  • Rental Servers & VPS
  • Travel

MiyaAILab

A hands-on lab for generative AI — new models, tools, and services tested for real, from benchmarks to everyday usefulness.

Lab

  • AI Lab トップ
  • 生成AI 全記事
  • ← Miya-Gadget 本体

Latest

  • Dual Tesla V100 SXM2 on a Single PCIe Slot: 64GB VRAM & 300 GB/s NVLink Tested — Is This $700 Setup Worth It?
  • Modded RTX 4080 32GB Benchmarked: Qwen3.8-27B at 262K Context, 125B MoE, and MiniMax H3 Video — What 32GB Actually Delivers
  • Tesla V100 32GB Runs Qwen3.8-27B: 131k Context on a Single Card — Measured Benchmark
  • Tesla V100 32GB in 2026: Local LLM Benchmark with Qwen 3.6 — 98.8 tok/s on MoE 35B, 1.6x Faster Than M1 Max (Used, ≈$900)
© 2026 Miya AI Lab — a section of Miya-Gadget. miyagadget.page