Verifies the currency and reliability of database tables by analyzing update timestamps and pipeline health.
The checking-freshness skill empowers Claude to autonomously validate the state of your data sources before performing critical analysis or engineering tasks. It systematically scans for common timestamp patterns, calculates data latency using SQL, and classifies table status from 'Fresh' to 'Very Stale.' By bridging the gap between database querying and Airflow pipeline monitoring, it ensures that your AI-driven workflows are always based on current, reliable information and helps diagnose root causes when data becomes stale.
Key Features
01SQL-based data age and latency calculations
02Standardized freshness status reporting (Fresh, Stale, Very Stale)
03Historical row count analysis to verify update consistency
04Automated timestamp column discovery across various ETL patterns
05Direct integration with Airflow DAGs for pipeline troubleshooting
0677 GitHub stars
Use Cases
01Confirming data ingestion status after a scheduled batch job
02Validating if a table is ready for reporting before generating insights
03Detecting silent ETL failures by monitoring data lag over time
