AI Blocks › Track your numbers
New dataset explorer
Give it a new dataset. Get its structure, spread, links and odd spots mapped before you use it.
New dataset explorer is a free AI skill for the job "Track your numbers". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · MIT licence · Made by Burhan Sebin
Made by Burhan Sebin. See the original on GitHub; we copied it as it is, with its licence.
What it does
A first look at a dataset you have not worked with, so surprises show up before they wreck a report or a model. You get a plain picture of what is really in the file. The skill reports size, column types and the first and last rows to catch bad headers or junk at the bottom. It counts missing values, constant columns and duplicates. Then it looks at each column's spread, such as average, median and skew, and flags lopsided categories. It checks which columns move together, flagging strong links, and finds outliers using the IQR method. It ends with a short summary of what you can trust and what needs cleaning first.
You give it
- A table of data (CSV, Excel or similar)
- The question you hope to answer (optional)
- Which column you want to predict (optional)
You get
- Size, types and sample rows
- A missing data and duplicate report
- Charts of each column's spread
- Strong links between columns
- Outliers and a short findings summary
Use it when
- A colleague sends a spreadsheet and you do not know what is in it.
- You are about to build a model and want to check the data first.
- Your totals look wrong and you suspect bad rows.
- You need a quick profile for a data handover.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run code and open spreadsheets.