Schema management

The Schema tab lets you explore your database structure, manage table relationships, and control which tables the AI can access. It offers two viewing modes and tools for defining semantic relationships between tables.

Schema scanning

When you first create a connection, Kaelio runs an initial schema scan to discover tables and columns. You can also trigger scans manually or set up automatic schedules (see Connection Settings).

While a scan is running, a progress indicator appears at the top of the Schema tab. The scan imports table names, column names, data types, and constraints.

Table view

The default view displays your schema as a structured table list. For each table, you can see:

Table name and column count
Column names, data types, and constraints
Whether the table is enabled for AI queries

Use the search bar to filter tables and columns by name.

Graph view

Toggle to the graph view using the icon in the top-right corner. This displays an interactive diagram showing:

Nodes — Each table appears as a node listing its columns
Edges — Lines connecting columns that have defined relationships
Navigation — Drag to pan, scroll to zoom, and click nodes to select them

Search for specific tables or columns, and the graph highlights matching elements.

Managing relationships

Relationships (also called links) define how columns in different tables relate to each other. These help Kaelio write more accurate JOIN queries.

Creating a relationship

In graph view, click Create Link or use the relationship creation button.

Select the source table and column.

Select the target table and column.

Optionally add a comment describing the relationship (for example, "Foreign key: orders belong to customers").

Save the link. It appears immediately as an edge in the graph.

Editing and deleting relationships

Click an existing relationship edge to view its details. From there you can update the comment or delete the relationship entirely.

Relationships help the AI understand how your tables connect. Well-defined relationships lead to more accurate JOIN clauses in generated SQL.

Connection Settings — Configure scan schedules and manage schema metadata
SQL Security Policies — Control which tables and columns users can access