AI Agent Use Cases
Views —

AI Agent Use Cases

Added today

44All 189

Prompt to Prepare a Shopping Cart from a Recipe Photo

The author reports that Grok Bot receives a recipe photo, organizes the ingredients and quantities, and creates a shopping list by excluding ingredients already available at home.

Grok Bot
Life & Hobbies

1–20 of 189

Life & HobbiesWalkthrough

Creating an Excel Household Ledger with Formulas and Input Rules

The author reports that by describing the required tables for Claude's Excel demo, the system reads files and executes Python code to create multiple sheets and formulas.

Claude

How it works

  1. The public Excel Demo is a local app built with Electron. After preparing the API key and execution environment, the author provides the desired table or an existing file.
  2. Claude is connected to tools for reading/writing files and executing commands, allowing it to create tables. The documentation gives examples such as a budget table and an exercise log.
  3. The author checks the generated file for sheet structure, formulas, formatting, and input validation. LibreOffice for formula recalculation is mentioned as optional.

Results

This is a local demo that creates an Excel file with multiple sheets and calculation formulas. The completed file is reviewed by the user after creation.

Notes

The creator specifies that this is a demo for local development and is not suitable for operational services or large-scale use. The author did not run it with actual household ledger data.

Source

GitHub Original post ↗

Example prompt · Claude

Goal: Create an Excel household ledger with formulas and input rules.

Inputs/Environment: [Anonymized income/expense table], [Category list], [Monthly budget], [Local Node.js/Python environment].

Reference: https://github.com/anthropics/claude-agent-sdk-demos/tree/main/excel-demo

Prepare to run the Excel Demo locally and create a workbook that compares monthly budgets with actual expenses. Keep the original file as read-only, and organize transaction details, category totals, and budget differences in a new file. Insert formulas where calculations are needed, and set input rules for date, amount, and category. Provide the Python code and file paths used for generation.

Do not speculate about account connection methods or outcomes not in the original documentation; first, list the required setup and execution steps. Start from a working copy to preserve existing data, and mark any items that cannot be accessed as needing verification.

Verification: Enter an empty transaction list, negative amounts for refunds, and multiple transactions in the same month, then compare the totals. Check for formula errors and incorrect references, and mark any cells that could not be recalculated separately.

With the results, organize the inputs used, execution logs, and findings. Clearly distinguish between completed tasks and those that failed or have not yet been executed.