dataflow

What problem does it solve?

DataFlow automates database CRUD and query workflows by generating workflow nodes directly from database models, so you stop writing repetitive SQL/ORM glue and start executing consistent, parameterized operations across SQL, MongoDB, and vector search.

Core Features & Use Cases

• Zero-config model-to-workflow node generation: Automatically creates 11 workflow nodes per SQL model (and a smaller set for MongoDB/vector use cases), covering create/read/update/delete/list/upsert/count plus bulk variants.
• Multi-database + multi-tenant patterns: Supports PostgreSQL, MySQL, SQLite, MongoDB, and pgvector-style vector operations, with enterprise-grade tenant isolation and per-instance isolation.
• Operational safety & DX guardrails: Provides structured validation rules (including strict parameter handling), safe string-ID preservation, and improved debugging via DF-XXX error codes and inspector patterns.
• Use cases: Build CRUD APIs (optionally via Express-style direct invocation), implement migrations and bulk ingestion, run transactional multi-step workflows (when you explicitly opt into TransactionScopeNode), and integrate with Nexus and Fabric Engine for external data products.

Quick Start

Ask for: set up a DataFlow instance for SQLite, define a User model with @db.model, and run an express create and read flow using await db.express.create and await db.express.read.