Pull names from a Wikipedia article and save them

Claude Computer Use opened Firefox, scanned a Wikipedia article, and wrote the names it found into a text file.

Done with Claude computer use· Beginner· Apr 2025

What they did
The user ran Claude Computer Use locally with Anthropic’s Docker image and their API key. They asked it to use Firefox to reach Wikipedia, look up a topic, identify the names on the page, and put those names into a text file.
How it went
The run took about a minute and seemed to handle the extraction correctly. Its biggest downside was price: the job burned nearly 100,000 tokens and cost around $0.31, which the poster judged too much for casual use.
Tips
This is a decent fit for a short, structured page, but token usage can be surprisingly high even when the task is quick. If you try it, keep an eye on cost before relying on it for repeat work.

Try it yourself with Claude computer use

Launch [browser] and go to the Wikipedia article about [topic]. Read the page, gather each person’s name mentioned there, and save the unique names as separate lines in [filename] inside [folder]. Before any saving, overwriting, submitting, sending, deleting, or other irreversible action, stop and ask me to approve the exact list. Once I approve, write the file and report the total number of names found.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.