Exploratory Data Analysis

What it does

This skill enables AI agents to perform exploratory data analysis (EDA) on datasets, helping users understand underlying patterns and distributions. It automates the initial investigation phase of data science workflows to reveal insights before deeper modeling occurs.

When to use it

• You need to quickly summarize the structure and content of a raw dataset.
• You want to identify missing values, outliers, or unusual data distributions automatically.
• You are preparing data for machine learning models and require statistical summaries.
• You need to visualize correlations between variables without writing complex code.

Key capabilities

• Automatically generates descriptive statistics (mean, median, mode, standard deviation).
• Detects and reports missing data points within columns.
• Identifies potential outliers based on statistical thresholds.
• Creates visualizations to show data distributions and relationships.
• Provides a high-level overview of dataset dimensions and types.

Example prompts

• "Run an exploratory data analysis on this CSV file containing customer sales records."
• "What are the key statistics and any missing values in this dataset?"
• "Generate a summary report including outliers and correlations for the provided JSON data."

Tips & gotchas

Ensure your input data is clean or pre-formatted, as EDA tools may struggle with unstructured text files. This skill focuses on discovery; you will still need to manually decide which variables to include in final predictive models based on the generated insights.