eda

Files (1)

• SKILL.md

SKILL.md (first lines)

# Exploratory Data Analysis

Run a structured first-pass EDA on a dataset: shape, types, missingness, distributions, and relationships.

## When to use
Trigger when the user is doing data science / ML work touching Exploratory Data Analysis on a dataset or notebook in this project.

## Steps
1. Load the data and report shape, dtypes, and memory.
2. Quantify missing values and duplicates.
3. Summarize numeric (describe) and categorical (value_counts) features.
4. Surface correlations and obvious data-quality issues, then list hypotheses to test.

## Output
Produce runnable Python (pandas / scikit-learn) plus a short written read-out of findings
and decisions. Prefer pipelines over ad-hoc steps, and always state assumptions.

From kakamana/ds-agent-starter, MIT.

More from kakamana/ds-agent-starter