
Point this at two SQL Server instances, say prod and staging, and it walks each table, detects primary keys from metadata, diffs the schemas, and reports missing rows, extras, and column mismatches per table. Under the hood it pulls data through mssql-python with Arrow via cursor.arrow(), and for big tables it hashes rows with HASHBYTES first so only the mismatched ones get fetched in full. It's a solid fit for post-migration validation or checking whether an ETL run actually landed. The honest caveat: it assumes you can hit both servers from one machine and that your PKs are discoverable, so it stalls if the schema is messy or the network is locked down. Worth reading the normalization rules before trusting a "identical" result on decimals and datetimes.
$ npx -y skills add github/awesome-copilot --skill sql-server-table-reconciliation --agent claude-codeInstalls into .claude/skills of the current project.
Compare identical tables across two SQL Server instances using Python with mssql-python driver and Apache Arrow. Detect missing rows, column mismatches, schema drift, and produce a reconciliation report.
Sponsors