← Back to library
MarketingPrompt

Gather the X, Reddit and LinkedIn debate into one sheet

Before each article, Daria Cupareanu used to lose an hour checking what people were saying on X, Reddit and LinkedIn. Sent from her phone to Claude driving a browser, she now gets a research sheet from those three sites in one pass.

  • OutcomeGet a single spreadsheet of the most discussed posts on a topic across three platforms, with a summary and link for each, before you start writing.
  • SetupMinutes
  • LevelStarter
  • CreditDaria Cupareanu
View original
Works with
Claude CoworkClaudeChatGPT agent

Published Oct 23, 2026Updated Oct 23, 2026AI dry run on sample input: passed Oct 23, 2026Source verified Oct 24, 2026

When should you use it?

  • You write regularly and research the conversation first
  • You want a sourced list rather than a vague summary
  • You can leave your computer signed in to the sites

Not for: Anything needing exact engagement metrics for reporting; Accounts you are not allowed to automate

What do you need, and what do you get?

You need

  • Topic and keywords
  • Time window, for example the last two weeks
  • The columns you want in the sheet

You get

  • Excel file with platform, author, summary, engagement and link
  • A second tab of recurring themes

How does it work, step by step?

  1. Name the topic and windowSay what the piece is about and how recent the posts must be.
  2. Fix the columnsList the fields you want per post so the output is consistent every time.
  3. Run it while you are awaySend the request from your phone if your setup allows remote use of the computer.
  4. Spot-check linksOpen a handful of rows to confirm the summary matches the post.
  5. Fill missing numbersExpect some platforms to hide counts from the agent; note them as unknown rather than guessing.

The prompt

The original prompt by Daria Cupareanu, word for word, with a link to where it was published. Below it is our adapted version with fill-in blanks, if you want it. Works in Claude Cowork, Claude and ChatGPT agent. Values in lime are placeholders.

Original prompt by Daria Cupareanu

Word for wordby Daria Cupareanu
I'm writing an article about Claude computer use and how non-technical people can use it. I need you to research what people are discussing about this topic across three platforms: X (Twitter), Reddit, and LinkedIn. For each platform:
- Search for the topic and related keywords
- Find the 10-15 most interesting or high-engagement posts from the last 2 weeks
- Note what questions people are asking, what angles they're covering, and what's getting the most engagement Put everything in an Excel spreadsheet with these columns:
- Platform (X, Reddit, or LinkedIn)
- Author/username
- Post content summary (2-3 sentences)
- Engagement (likes, comments, reposts/upvotes)
- Link to the original post
- Key takeaway (what this post adds to the conversation) Save the file to my Desktop as "content-research-[topic].xlsx"

Exactly as published by Daria Cupareanu in Claude Computer Use: 6 Real Workflows Tested.

Our adapted version (optional)

Adapted prompt0 variables
Look up recent discussion of [topic] from the past [two weeks]. Cover three sites (X, Reddit, LinkedIn); on each, pick the 10 to 15 posts with the most discussion. Build a spreadsheet with: platform, author, two-sentence summary, engagement figures if visible (write unknown if not), link and the one idea each post adds. Add a second tab listing the recurring themes and open questions. Save it as research-[topic].xlsx and tell me which sites hid engagement data.

What can go wrong?

  • Counts may be missing on some sites, as the writer saw with X and Reddit.
  • Without a date window you get stale viral posts instead of the current debate.

Our take

A tidy, unglamorous win and the right kind of first job for a computer-use agent: read-only, low stakes and easy to check. Fixed columns are what make the sheet useful week after week; the same fields per post let you compare debates across topics. Set a date window every time, or the agent will happily surface old viral posts instead of the current conversation. Daria found some engagement counts hidden from the agent, so mark those unknown rather than letting it guess, and do not use the sheet for metric reporting. Open a handful of rows and confirm the summary matches the post before quoting anyone. Running it remotely while you are away is convenient, but it means leaving a signed-in computer for the agent to drive, so only automate accounts you are allowed to and keep it to reading. It never posts or replies as you. For a writer, saving an hour of tab-hopping per article adds up quickly.

Where the time goes back: An hour of tab-hopping becomes a ten-minute background job with a tidier result than you would make by hand.

FAQ

How long did it take?
About ten minutes for 44 posts in the writer's run.
Can it reach every site?
Only those already signed in on the computer it controls, and some hide engagement data.

Source and credit

The original prompt on this page is by Daria Cupareanu, shown word for word with a link to where it was published. The summary, notes and adapted version are ours.

Original
Claude Computer Use: 6 Real Workflows Tested
Creator
Daria Cupareanu
Published on
aiblewmymind.substack.com (Substack)
Original date
Apr 2, 2026
License
Not openly licensed; the original prompt is shown word for word with credit and a link to the original
Originally for
Claude Cowork
AI dry run
Oct 23, 2026: passed. Dry-run tested by our AI test agent on sample input. PASS: Gates held on desk check; no live sends or purchases.
AI dry run on sample data; not run with live connected accounts. The playbooks.bot test agent filled the prompt with synthetic inputs, produced the output and checked it against the page's stated outputs and guardrails.
Verified
Oct 24, 2026: source link resolves to the original, the credited creator published it, our excerpt or prompt matches it, and the license allows our use.

Are you the creator? Claim or correct this page.

More in Marketing

All 97 Marketing playbooks →

Related playbooks

MarketingPrompt✓ AI dry run

Answer a journalist's request using only facts you can back

For each media request, find out quickly whether you are a fit, then get a short answer-first draft built only from experience and data you confirmed, with unsupported lines flagged and the deadline checked.

Credit: Kourosh Ghaffari (@BuildingWaves)

Minutes
MarketingWorkflow✓ AI dry run

Build a weekly LinkedIn Ads pacing report with Claude

Have Claude turn each week's LinkedIn Ads results into one written report: progress against pipeline targets, the main platform metrics, the audiences, campaigns and ads that are falling behind, ads whose click-through to the landing page is slipping, and a short list of last week's finished work and what is planned next.

Credit: Patrick Cumming

An afternoon+
MarketingPrompt✓ AI dry run

Recover backlinks that point at pages you moved

Get a reviewed change list of redirects or restorations for old URLs that other sites still link to, plus a recheck plan, so those visitors and links reach a page that serves the same purpose again.

Credit: Kourosh Ghaffari (@BuildingWaves)

Under an hour
MarketingRoutine✓ AI dry run

Review which quote tweets worked and tune the scan

Every week, see which scan signals actually predicted your strongest quote tweets, with proposed rule changes you approve before the next morning scan uses them.

Credit: James Camp (@JamesonCamp)

Minutes