data-explore

Profile tabular datasets with reports, visual diagnostics, and outlier detection.

Updated Mar 3, 2026

Search Tags:#profiling #exploratory data analysis #eda #data-quality #outlier-detection #timeseries-analysis

Ready to run

npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-explore

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: data-explore
Source: https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace/tree/main/plugins/data-analysis/skills/data-explore
Command: npx skills add https://github.com/mutsumi-yamamoto/claude-data-analysis-marketplace --skill data-explore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill performs senior-level exploratory data analysis to quickly identify data quality issues, distributional properties, correlations, outliers, and time-series characteristics so analysts can decide data-cleaning and modeling priorities.

Core Features & Use Cases

Automated profiling: Generates comprehensive profiling reports and extended summary statistics including percentiles, skewness, kurtosis, and normality tests.
Visualization & diagnostics: Produces histograms, KDE, QQ plots, boxplots, correlation heatmaps, missingness matrices, STL decomposition, ACF/PACF plots, and target distribution visualizations.
Outlier & missingness analysis: Detects univariate and multivariate outliers (IQR, IsolationForest, LOF), assesses missingness patterns (MCAR/MAR/MNAR), computes mutual information and Cramér's V for categorical relationships.

Quick Start

Run a senior-level exploratory data analysis on the project's dataset and save the generated reports and visualizations to the data/docs/ folder and record the summary in analysis_context.md.

Frequently Asked Questions about data-explore

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I perform exploratory data analysis on a CSV dataset to identify data quality issues?▼

Exploratory data analysis on CSV datasets generates profiling reports, missingness matrices, and outlier detections to identify data quality issues. It computes extended summary statistics including skewness, kurtosis, and normality tests to guide cleaning priorities.

How can I detect univariate and multivariate outliers in a mixed-type tabular dataset?▼

Outlier detection in mixed-type datasets applies IQR, IsolationForest, and LOF algorithms to identify univariate and multivariate anomalies. It evaluates distributional properties and flags data points requiring review during feature selection.

What is the best way to assess missing data patterns and determine if they are MCAR, MAR, or MNAR?▼

Missingness assessment evaluates missing data patterns to classify them as MCAR, MAR, or MNAR. It generates missingness matrices and computes mutual information and Cramér's V to reveal relationships between missing values and categorical features.

Can I generate time-series diagnostics like STL decomposition and ACF/PACF plots during initial data profiling?▼

Time-series diagnostics during data profiling produce STL decomposition, ACF, and PACF plots to uncover temporal characteristics. These visual outputs are saved alongside standard distribution plots to support time-series feature evaluation.

Do I need specific Python libraries installed to run automated data profiling and generate visual diagnostics?▼

Automated data profiling requires Python data science libraries including pandas, numpy, scipy, seaborn, statsmodels, and scikit-learn. Optional ydata-profiling and missingno packages enable richer automated reports and missingness visualizations.

What visual diagnostics are generated for target variable analysis in a dataset?▼

Target variable analysis generates histograms, KDE plots, QQ plots, and boxplots to visualize target distributions. It pairs these with correlation heatmaps to highlight relationships between features and the target during initial project discovery.