First public release of Test Data Workbench.
Point it at a database, get realistic, relationally intact test data back. The Workbench reflects an unknown schema, classifies each table, and emits readable per-table Python generators, dependency-ordered and referentially consistent.
Highlights:
- Schema-adaptive generation with a four-level fallback ladder (degrade, never fail)
- Deterministic output with --seed (byte-identical across runs)
- Strict --metadata-only analysis mode (zero SQL against row data)
- Name-based PII column flagging
- Post-generation constraint check (PK uniqueness, not-null, FK integrity)
- tdw CLI plus a FastAPI REST interface
- SQLite and PostgreSQL supported
Docs: https://volanticsystems.github.io/test-data-workbench/