Collect weekly chart data and get a report on who moved
Idea + starter prompt, not run by us yet
For: Newsletter writers and analysts who track a public leaderboard
Frank Andrade has Claude in Chrome collect the top 250 rows of a public Substack chart each Tuesday into a CSV, with rank, newsletter, date and free or paid. Later that day a Cowork task opens the saved files and makes a sheet listing the biggest risers, fallers and newcomers, split by paid and free. Collecting the data needs only a few minutes per week.
Steps
- Pick the public chart you want to follow and the columns you need.
- Run the collection once by hand and check the CSV.
- Turn it into a weekly scheduled task.
- Add a second task later the same day that compares this week with last week.
- Keep the computer awake with the Claude app open, since scheduled runs need it.
Starter prompt
Each week, read the folder [folder] that holds weekly CSV files of [chart or leaderboard name]. Compare the newest file with the one before. Build a spreadsheet with the biggest climbers, the biggest drops and new entries, separately for [group A] and [group B]. Add a short summary at the top. Do not delete or change the original CSV files.
Original post: artificialcorner.com · found via agentusecases.ai. Written up in our own words. We haven't run it, so test it on made-up data first.