Hooks up a DuckDB database file so you can query it with the companion query skill. It resolves the path, validates the file, explores the schema (tables, columns, row counts), and writes everything to a shared state.sql init file that persists the session. The state file lives either in your project directory or home directory, your choice, and it's append-only so multiple databases and other skills can coexist. Worth noting it handles the entire discovery flow automatically, from checking if DuckDB is installed to asking if you want to gitignore the state directory. If you're jumping between DuckDB files or setting up a new analytics workflow, this saves you from writing the same ATTACH boilerplate every time.
npx skills add https://github.com/duckdb/duckdb-skills --skill attach-db