Harvard Artifacts Etl Streamlit

This is a complete ETL reference implementation that pulls artifact data from Harvard's Art Museums API and pipes it into a relational database with a Streamlit frontend for exploration. You get schema definitions for storing metadata, media URLs, and color analysis across three normalized tables, plus code for paginated extraction, transformation logic that flattens nested JSON, and batch loading with proper foreign keys. The analytics layer includes pre-built SQL queries for slicing collections by culture, century, department, and dominant colors, all visualized through Plotly charts. It's built on MySQL but the patterns translate to any relational database. Good starting point if you're learning data pipeline architecture or need to stand up a quick proof-of-concept dashboard around any public API with similar structure.

Install to Claude Code

npx -y skills add aradotso/data-skills --skill harvard-artifacts-etl-streamlit --agent claude-code