Import from Structured Data

The structured import flows in Studio convert source text into editable diagrams. This is useful when you already have SQL, OpenAPI, Terraform, code, or infrastructure-oriented content and want a visual model quickly.

How it works

Paste your source text, choose the matching mode, and generate the diagram. Depending on the input type, OpenFlowKit either runs a specialized AI-assisted import or a deterministic parser-based flow.

One common pipeline is:

Your text → specialized import flow → DSL or graph model → layout → canvas

All your existing AI provider and API key settings apply.

For fully deterministic infrastructure parsing that does not need AI, use Infrastructure Sync.

SQL DDL → Entity-Relationship Diagram

Paste one or more CREATE TABLE statements to generate an ER diagram.

Flowpilot will:

• create one node per table
• list primary keys, foreign keys, and key columns inside each node
• draw edges between tables that share foreign key relationships
• group related tables where the model can infer a useful structure

OpenAPI and service-structure imports

Use OpenAPI import when you want a service or endpoint-level first pass from an existing API spec.

Code and architecture imports

Use code-oriented import paths when you want a draft that reflects application structure instead of hand-drawing from scratch. This is a good fit for:

• backend service relationships
• module or package boundaries
• API-to-database handoffs
• architecture review prep before cleanup on the canvas

These imports are best treated as editable first drafts, not perfect final diagrams.

When to choose this flow

Choose structured import when a source artifact already exists and you want a fast editable draft. Choose manual editing when the diagram is primarily conceptual. Choose Infrastructure Sync when determinism matters more than AI interpretation.