EDA Report: Instant Dataset Profiling With Charts

    1

    Makes your AI agent profile any CSV, Parquet or Excel file and save a readable markdown EDA report with charts: types, missing values, distributions, correlations and the target. Never changes your data.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    EDA Report: Instant Dataset Profiling With Charts

    Example session with this skill installed

    Explore titanic3.csv for me. The target is survived.

    • Read your context and instructions
    • Compiled the eda report
    • Generated the data export

    Summary

    titanic3.csv has 1,309 rows and 14 columns, with 21% of cells missing. Three columns (cabin, boat, body) are mostly empty, and boat/body look like they are filled in only after the outcome: run leakage-check before modelling.

    1. Overview

    RowsColumnsMissing cellsDuplicate rowsMemory
    1,3091421.04%00.22 MB

    Column kinds: 6 numeric, 6 categorical, 1 boolean, 1 id

    2. Things to look at next

    • name looks like an ID (unique per row): do not use it as a feature.
    • fare is very skewed (skew 4.4): consider a log transform or a robust model.
    • ticket has 929 categories: one-hot encoding will create many columns; group rare values or use target encoding inside cross-validation.
    • cabin is 77% missing. Ask why it is missing: if it is only filled in for some outcomes it can leak the target (run leakage-check).
    • boat is 63% missing. Ask why it is missing (run leakage-check).
    • body is 91% missing. Ask why it is missing (run leakage-check).

    3. Columns (excerpt)

    ColumnKindTypeMissingUnique
    pclassnumericint640.0%3
    survivedbooleanint640.0%2
    nameidstr0.0%1,307
    sexcategoricalstr0.0%2
    agenumericfloat6420.09%98
    farenumericfloat640.08%281
    cabincategoricalstr77.46%186
    boatcategoricalstr62.87%27
    bodynumericfloat6490.76%121

    Also saved: eda_summary.json and charts in figures/ (missing values, distributions, top categories, correlations, target, target relationships).

    Next steps

    • Run data-cleaning for format problems.
    • Run leakage-check before training any model.

    titanic3.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Give your AI agent a proper first look at any dataset. Point it at a CSV, Parquet or Excel file and get a readable markdown EDA report with charts, without the agent touching your data.

    The report covers:

    • Shape, column types, and columns read as the wrong type (numbers stored as text, IDs read as numbers)
    • Missing values, with a chart
    • Numeric distributions and top categorical values
    • Correlations, and how each column relates to the target
    • A "Things to look at next" list, so you know what to fix before modelling

    It also teaches the agent good habits: explore training data only, describe without fixing, and treat a "too good" column as a warning, not a win.

    • Works with Claude Code and any agent that reads SKILL.md
    • One tested Python script: pandas, numpy, matplotlib
    • Large files are sampled automatically (200,000 rows by default)
    • No API keys, no network access
    • Open source (MIT): github.com/m-shahmeer-kz/dataskills

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Data & Databases.

    Files above 200,000 rows are profiled on a random sample (adjustable with --max-rows). Tabular files only: CSV, Parquet and Excel. Describes the data; it does not clean or change it.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 days ago

    • Free with an Agensi account

    Software
    Python 3.9+pip

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed2 days ago

    What's inside

    Frequently Asked Questions

    An overview table, a "Things to look at next" list, a column table with types and missing values, numeric and categorical summaries, correlations and a target section. Charts: missing values, numeric distributions, top categories, a correlation heatmap and target charts.

    This skill is compatible with any AI agent capable of running Python scripts and handles standard tabular formats including CSV, Parquet, and Excel (.xlsx).

    Files larger than 200,000 rows are profiled on a random sample of 200,000 rows to stay fast. You can change this with the --max-rows option.

    The numbers come from a tested Python script (pandas, numpy, matplotlib), not from the model guessing, so the statistics are computed directly from your file and the report is the same every time you run it.

    The report identifies data quality issues like format mismatches and missing values, but it is an observation tool only; it does not modify or clean your original files.

    The SKILL.md instructions, the tested eda_report.py script, requirements.txt and a short install note. Each run writes eda_report.md, eda_summary.json and the charts. The skill is free and open source (MIT).

    New in Data & Databases

    Free