data-analyst

Performs statistical analysis and exploratory data analysis to generate insights from datasets.

4|2|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/Arete-Consortium/ai-skills --skill data-analyst-arete-consortium
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: data-analyst
Source: https://github.com/Arete-Consortium/ai-skills/tree/main/personas/data/data-analyst
Command: npx skills add https://github.com/Arete-Consortium/ai-skills --skill data-analyst-arete-consortium

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, scipy, scikit-learn.

What problem does it solve? Raw datasets require rigorous statistical treatment before they can inform decisions, and ad-hoc analysis often skips assumption checks, misreads correlations, or reports results without uncertainty measures. This Skill turns Claude into a data analysis agent that applies proper statistical methods and produces reproducible, well-documented findings. ## Core Features & Use Cases - Exploratory Data Analysis: Profiles dataset shape, types, distributions, missing values, and data quality with structured summary output. - Hypothesis Testing: Selects appropriate tests (t-test vs Mann-Whitney U) based on normality checks, and reports p-values, effect sizes, and confidence intervals. - Trend & Correlation Analysis: Decomposes time series into trend, seasonality, and residuals, and ranks variable correlations with significance flags. - Use Case: Given a new CSV of product usage metrics, ask for an EDA to surface data quality issues, identify which features correlate with churn, and test whether two user segments differ significantly. ## Quick Start Ask Claude to perform an exploratory data analysis on your dataset and test whether the differences between two groups are statistically significant.

Frequently Asked Questions about data-analyst

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

FAQPage Schema
How do I perform exploratory data analysis on a new dataset?▼

Start by profiling rows, columns, data types, and missing value percentages, then examine distributions with pandas describe() and per-column mean and standard deviation. This Skill generates that EDA workflow automatically with a structured data quality table and key findings.

How to choose between t-test and Mann-Whitney U test in Python?▼

Check normality of both groups with the Shapiro-Wilk test first. If both groups are normal (p > 0.05), use scipy's ttest_ind; otherwise use mannwhitneyu. This Skill implements that decision logic automatically and reports the test used, statistic, and p-value.

What Python libraries are used for statistical analysis here?▼

The analysis code uses pandas for data manipulation, numpy for numerical operations, scipy.stats for hypothesis tests and correlations, and scikit-learn for modeling tasks. All code is provided in complete, reproducible form.

Why does correlation not imply causation in data analysis?▼

Confounding variables frequently produce spurious correlations, so a significant Pearson coefficient alone cannot support causal claims. Causal conclusions require experimental design or controlled studies, which is why this Skill reports correlations with significance flags but avoids causal language.

When should I not use this data analysis skill?▼

Avoid it for building data ingestion pipelines (use a data engineering skill), creating charts as the primary deliverable (use a visualization skill), or writing executive stakeholder reports (use a report generation skill). It focuses on statistical rigor, not infrastructure or presentation.