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

Dev

Checking Accessibility with a Browser and Saving Execution Evidence

The author reports a usage record where a coding agent ran a browser and inspection tools to check accessibility, leaving behind both the results and screen evidence.

Claude Code

How it works

  1. The author prepared the web page to be checked, the Showboat and Rodney environments, and determined the screen flow to be verified.
  2. There is an example where the coding agent executed browser control and accessibility inspection commands, then documented the results and screenshots.

Results

The author published an example where the agent executed accessibility inspection commands and documented both the screen and results.

Notes

This is an accessibility inspection usage record published by the author. Showboat and Rodney are auxiliary tools called by the agent, and this article focuses on using Claude Code.

Source

Simon Willison의 Showboat·Rodney 사용 기록 Original post ↗

Example prompt · Claude Code

Goal: Check accessibility with a browser and save execution evidence.

Input & Environment: [Test page], [Inspection flow], [Showboat·Rodney environment], [Result storage location].

Reference original: https://simonwillison.net/2026/Feb/10/showboat-and-rodney/

Based on the published workflow, prepare an implementation or work procedure that can be followed in a test environment. Do not guess internal code or settings not disclosed in the original; if alternative connections are needed, indicate so.

Connections and materials to prepare: Prepare the web page to be checked, the Showboat and Rodney environments, and determine the screen flow to be verified.

Agent actions to implement: There is an example where the coding agent executes browser control and accessibility inspection commands, then documents the results and screenshots.

Scope of actions: This is an accessibility inspection usage record published by the author. Showboat and Rodney are auxiliary tools called by the agent, and this article focuses on using Claude Code.

The deliverable is a document containing the accessibility inspection results and execution evidence. Also organize the necessary setup steps and actual execution methods. After checking the results returned by the tool, select the next task, and separately note any inaccessible materials or tasks that could not be executed.

Verification: Check whether the tool execution record and screenshots refer to the same screen, and whether keyboard navigation and input names were confirmed on the actual screen. Clearly distinguish the inputs used for verification, the actual outputs, and any items not yet confirmed.