skills.rest
zebbern
zebbernCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

dataset-quality-audit

Audit tabular data across 12 quality dimensions and produce scores, grades, and fix suggestions.

4.6k|462|Updated Jun 21, 2025
Search Tags:#audit#Pandas#CSV#missing values#outlier detection#Data Quality#Data Profiling
One-click install
npx skills add https://github.com/zebbern/claude-code-guide --skill dataset-quality-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dataset-quality-audit
Source: https://github.com/zebbern/claude-code-guide/tree/main/skills/dataset-quality-audit
Command: npx skills add https://github.com/zebbern/claude-code-guide --skill dataset-quality-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, and includes scripts (resource) components.

What problem does it solve?

Tabular datasets often contain hidden issues like missing values, duplicate rows, outliers, inconsistent formats, and mixed types that silently corrupt analysis. This Skill automates a comprehensive quality audit of CSV, TSV, Excel, and JSON files, producing a scored report with actionable remediation suggestions instead of requiring manual inspection.

Core Features & Use Cases

  • 12-Dimension Quality Checks: Detects missing values, duplicates, type inconsistencies, IQR outliers, format violations (email/phone/date/URL), uniqueness breaches, whitespace issues, constant columns, skewed distributions, naming problems, cardinality anomalies, and cross-column logic violations.
  • Scored JSON Report: Outputs per-dimension scores (0-100), a weighted overall score, a letter grade (A+ to F), and a prioritized list of fix suggestions.
  • Use Case: Before training a model on a 10,000-row customer CSV, run the audit to discover that the age column has 1.5% missing values, 200 rows are fully duplicated, and the created_at column mixes date formats, then apply the suggested fixes.

Quick Start

Ask the assistant to run a data quality check on your CSV or Excel file, optionally specifying ID columns and date columns, and review the generated score, grade, and fix suggestions.

Frequently Asked Questions about dataset-quality-audit

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

FAQPage Schema
How do I check data quality of a CSV file in Python?▼

Run the data_quality_checker.py script with your CSV file path as the argument. It loads the file with pandas, runs 12 quality dimension checks, and prints a JSON report with scores, a grade, and fix suggestions to stdout.

What data quality issues can pandas detect automatically?▼

This tool detects missing values, duplicate rows, mixed column types, IQR-based outliers, invalid email/phone/date formats, duplicate IDs, whitespace problems, constant columns, skewed distributions, poor column naming, cardinality anomalies, and cross-column logic violations.

Does the data quality checker support Excel and JSON files?▼

Yes, it supports CSV, TSV, Excel (.xls/.xlsx), and JSON files, detected by file extension. Excel files are read with pandas read_excel, and JSON files with read_json, while CSV and TSV accept a configurable encoding parameter.

How do I check uniqueness of specific ID columns in a dataset?▼

Pass the --id-columns flag with comma-separated column names, such as --id-columns "user_id,email". Without it, the tool auto-detects ID-like columns by names containing id, uuid, or similar keywords.

Can I audit a large dataset without loading all rows?▼

Yes, use the --sample flag with a row count to randomly sample rows with a fixed random seed before running checks. This speeds up audits on large files while keeping results reproducible.

What are the limitations of IQR-based outlier detection?▼

IQR detection only applies to numeric columns with at least four values and flags values outside 1.5 times the interquartile range. It may flag legitimate extreme values in heavily skewed distributions, so results should be reviewed in domain context.

Related Skills

Explore More Data & Analytics Skills→
GayanKavindaGayanKavinda

xlsx

Create, edit, and repair spreadsheet files from raw tabular data.

Community
Advanced
brainbrewlabsbrainbrewlabs
7

data-cleaning

Profile and clean tabular datasets, then write a cleaned file with a change report.

Official
Intermediate
w95w95
145

data-exploration

Profile datasets to assess structure, quality, and column patterns.

Community
Intermediate
ArogyaReddyArogyaReddy

data-quality-frameworks

Automate data quality validation with Great Expectations, dbt tests, and data contracts.

Community
Advanced