data-quality-guardian

Specialist squad for data quality in data pipelines.

Overview

The data-quality-guardian is a complete squad covering the entire data quality pipeline:

1. Data Profiling — Distribution analysis, data types, cardinality, null rates, uniqueness, and descriptive statistics
2. Anomaly Detection — Outlier identification, data drift, unusual patterns, and baseline deviations
3. Schema Validation — Type checking, constraints, referential integrity, and breaking change detection
4. Quality Reports — Composite scores across 6 dimensions, trends, SLA comparisons, and executive summaries
5. Remediation Suggestions — Automated cleaning scripts, pipeline fixes, and governance recommendations

Pain Point: 89% of data teams report quality issues; 61% list data quality as the top challenge in their organizations.

Agents

Workflows

Available Commands

Quick Start

# Activate the quality reporter (main orchestrator)
/dqg:agents:data-quality-reporter

# Full data quality audit
*full-audit

# Quick data check
*quick-check

# Profiling only
*profile-data

# Anomaly detection only
*detect-anomalies

Target Users

• Data Engineers
• Data Analysts
• Data Governance Teams
• Analytics Engineers
• CTOs and Data Leaders

Requirements

• Access to the dataset for analysis (CSV, Parquet, JSON, SQL)
• Expected schema documented (optional, can be inferred)
• Previous baseline for comparison (optional)
• Quality SLA definitions (optional)