Exploratory data analysis
Skill minagayid/Agents/skills/library/exploratory-data-analysis
Run a structured exploratory data analysis (EDA) on a dataset to surface distributions, relationships, and issues. Use when the user asks to explore, understand, or summarize a dataset.From its SKILL.md
npx -y skills add minagayid/Agents --skill exploratory-data-analysisAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.5 KB, as published. Nobody here has run it
Exploratory Data Analysis
Understand a dataset before modeling or drawing conclusions.
Steps
- Frame the question. What decision or hypothesis is this analysis serving?
- Overview. Shape, dtypes, missingness, memory; identify id/target/feature roles.
- Univariate. Distribution of each variable — histograms/boxplots for numeric, value counts for categorical. Note skew, spread, outliers, rare categories.
- Bivariate / multivariate. Relationships to the target and among features: correlations, grouped stats, scatter/box by category. Watch for confounders.
- Quality issues. Missing patterns (random vs. systematic), inconsistent categories, leakage, class imbalance, duplicated or near-constant columns.
- Summarize findings with the visuals that carry the insight (see the
datavizguidance for good chart/color choices) and a short list of implications.
Guidelines
- Let the data surprise you — don't just confirm the expected story.
- Prefer clear visuals over tables of raw numbers; label axes and units.
- Distinguish correlation from causation explicitly.
- Record concrete next steps: features to engineer, data to fix, questions raised.
- Keep it reproducible (a notebook/script), seeded where randomness is involved.