Profile from CSV — Development

This skill audits datasets for data quality issues such as missing values, outliers, duplicates, and inconsistencies. It profiles data distributions, assigns a Data Quality Score (DQS), and produces a prioritized remediation plan to make data ready for analysis or modeling.

engineering claude-code

📂 View skill on GitHub ↗ 🚀 How to use

✨ What it does

Profiles datasets: shape, types, completeness, distributions, and DQS.
Analyzes missing value patterns and classifies them as MCAR/MAR/MNAR.
Detects outliers using IQR, Z-score, and modified Z-score methods.
Checks for duplicates, referential integrity, and logical constraints.
Generates prioritized remediation plans and monitoring thresholds.

🎯 When to use it

When you need to check the overall quality of a dataset before analysis or model training.
When you suspect specific issues like missing values, outliers, or duplicates in a dataset.
When you want to set up ongoing monitoring for data quality on a live pipeline.
When you need to compare current data distributions against a baseline to detect drift.

🚀 How to use

Trigger the skill by asking to audit data quality, profile a dataset, or find outliers/missing values. Provide the path to a CSV file. The skill runs Python scripts to analyze the data and produces a report. Example prompts:

Profile this dataset: data.csv
Check data quality of data.csv and give a remediation plan.

📄 Output: A data quality report with DQS score, per-column breakdown, and prioritized remediation steps.

📦 Add this skill to Claude Code

# 1. Get the skills repo
git clone --depth 1 https://github.com/alirezarezvani/claude-skills /tmp/claude-skills

# 2. Copy this skill into your project (or ~/.claude/skills for all projects)
mkdir -p .claude/skills
cp -r /tmp/claude-skills/engineering/data-quality-auditor/skills/data-quality-auditor .claude/skills/data-quality-auditor

Skill source: engineering/data-quality-auditor/skills/data-quality-auditor/SKILL.md

⚠️ Good to know

Requires Python environment with necessary libraries; assumes CSV input; domain knowledge may be needed for certain findings.

❓ FAQ

What is the Data Quality Score (DQS)?

DQS is a 0-100 composite score across five dimensions: completeness, consistency, validity, uniqueness, and timeliness, with weights 30%, 25%, 20%, 15%, and 10% respectively.

How does the skill handle missing values?

It classifies missingness patterns (MCAR/MAR/MNAR) and recommends imputation strategies based on null percentage, such as dropping rows for <1% or flagging for review if >30%.

Can I use this skill for monitoring?

Yes, use the --monitor flag with data_profiler.py to generate threshold-ready summaries for alerting on critical columns.

🤖 Overview, features, install steps and FAQ were generated from the project's SKILL.md on Sep 4, 2026. Always check the original source before running commands.