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

Work & Research

Creating Danggeun Cafe Automation via Voice Call

This is a report from South Korea about someone creating Danggeun Cafe automation by explaining tasks verbally while keeping a Grok Bot call active. The author reports they had been on the call for about five hours since the morning.

Grok Bot

How it works

  1. The author enabled the Grok Bot call feature and continued the conversation to set up Danggeun Cafe automation.

Results

The author said the call quality was clear. They did not disclose the results or any figures related to the automation.

Notes

Since this was written during the process, the scope and completion status of the automation are unknown. The author also mentioned that the usage of Grok Bot, Cursor, and Grok was combined.

Source

X Original post ↗

Example prompt · Grok Bot

Organize the Danggeun Cafe repetitive tasks you want using voice conversation, and start by creating small automations.

What you should explain: the current tasks [task], the screens you use [screen], the starting conditions [condition], the required input and desired outcome [content]. First, briefly summarize what you explained and confirm it with the user, then ask about any missing conditions one at a time. Do not guess about features or Danggeun Cafe integration methods that have not yet been decided.

Divide the work into input, processing, and result confirmation, and choose the smallest possible flow. First, investigate which connections, tools, and manual steps are available in the current environment, and implement only in ways that are permitted. If there is no way to connect, distinguish between what can be implemented and what is blocked.

After running a sample once with test data, explain the results and the next options so the user can modify them by voice. If external posting or message sending is included, prepare a draft for review first.

Deliverables: confirmed requirements, implementation files or settings, execution method and test results. Clearly distinguish between what was explained by voice, what was actually created, and what has not yet been verified to work.