Research web pages and drop screenshots into a document

A reviewer used Operator to gather web info and organize it in a Google Doc, but the agent struggled with browsing depth and could not handle screenshots.

Done with Operatorยท Intermediate

What they did
They gave Operator a week of plain browser tasks, including checking design posts, turning AI news into summaries, compiling website builders with prices, and collecting page screenshots into a shared Google Doc. Operator searched on its own, while the reviewer had to try again with fresh instructions when it drifted or got stuck. The reviewer also stepped in at logins and CAPTCHA screens.
How it went
The output was mostly poor: it repeated old tweets, summarized only one Axios piece, and pulled the first ten entries from a Bing list page instead of doing broader research. It also kept bouncing between tabs on multi-article jobs and never managed the screenshots.
Tips
Plan to supervise anything that may involve sign-ins, payments, or CAPTCHA checks. For multi-part research, ask for checkpoints so you can catch drift early.

Try it yourself with Operator

Help me research [topic] by visiting the main pages and related sources. For each item, record the site name, starting price, and 2-3 useful details, then add a homepage screenshot or note if you cannot capture one. Put the findings in a new Google Doc named [doc name]. Stop and ask me to take over before any login, payment, CAPTCHA, or other irreversible step, and wait for my approval before continuing.

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

Read the original on icons8.com โ†—

Get 3 agent uses like this every morning

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