Schemas

Browse a schema

The schema workspace is where you read and edit one schema. It is a single page with a header and a row of tabs, one per kind of object or task.

In the app: Data Model › Schemas › your schema

[screenshot]
A schema workspace on the Tables tab, with the header, version badge, Versions button and tab row visible.

The header

  • Labels as coloured chips, top left.
  • Current version, for example (current v2.0.1), or (no versions).
  • Versions opens the version history. See Browse the version history.
  • A metadata row with the schema’s dialect, who last modified it, and when it was modified and created.

The selected tab is kept in the page address, so you can bookmark or share a link straight to, say, the Diagram tab.

The tabs

Tab What it’s for
Info Name, description, database schema name, labels, custom attributes, table count, versioning, and the ACTIONS card (see below).
Tables Search, create, bulk-describe and bulk-delete tables. See Work with tables and columns.
Views, Procedures, Functions, Scripts Lists of SQL objects. See Work with views, procedures, functions and scripts.
Enums Named lists of allowed values. See Work with enums.
Diagram An interactive ER diagram of the whole schema. See Read a schema as an ER diagram.
Schema Agent Chat with an AI advisor that can propose new objects and changes. See Work with the Schema Agent.
Star Schema Kimball-style plans built on this schema. Shown only when your organization has the Star Schema feature. See Star schema overview.
Pipeline Shown only when this schema is the target of a Pipeline. Holds the Pipeline’s load settings. See Pipelines overview.
Attributes Custom attribute values on the schema itself. See Attach labels and attributes.
Governance Placeholder. Shows “Coming soon”.
Query Agent Describe a query in plain English and get SQL back. See Generate a query with AI.
MCP Server Share the schema’s metadata with AI assistants. See Expose a schema over MCP.
Settings Default dialect, naming conventions, environments, attribute groups, and Delete Schema. See Schema settings.
Users Admins only. Override a user’s role for this schema.

The Info tab

The Info tab has cards for:

  • The schema name, with Rename Schema.
  • DESCRIPTION, with an auto-generate button (AI) and an edit button.
  • Custom attributes, with a link to the Attributes tab.
  • DATABASE SCHEMA: the database schema name objects are created in, for example dbo. The reset button clears per-table and per-object overrides so everything uses this value.
  • LABELS, with an edit button.
  • TABLES: the table count.
  • VERSIONING: the latest version, Create Version and View All.
  • ACTIONS:

Import descriptions

Import Descriptions fills in descriptions for tables, columns, views, stored procedures, functions and parameters that already exist in the schema.

  1. Upload a .sql, .txt, .csv, .json, .yaml or .yml file (up to 5 MB), or paste text under Or Paste Text.
  2. Pick a mode. By default, only objects with no description are updated. Tick Overwrite existing descriptions to replace every matched description, or Append to existing descriptions to add the new text after the old on a new line.
  3. Click Import.

The import runs in the background; you can close the dialog while it works.

Bulk actions on the object lists

The Tables, Views, Procedures, Functions, Scripts and Enums lists share the same pattern:

  1. Tick rows, or tick the header box to select every row on the page. Unticking the header box clears every selection.
  2. Click Generate Descriptions to have AI write descriptions for the selected items. Existing descriptions are overwritten, and this can’t be undone. The work runs in the background.
  3. Or click Delete … to delete the selected items. Kenseme blocks the delete if other objects depend on them and lists the dependencies.

Grid columns can be resized by dragging their edges. The Tables tab also has a Per page selector (10, 20, 50 or 100).

Descriptions

Most objects (the schema, tables, columns, views, procedures, functions, enums) have a description. You can type one, click the auto-generate button to have AI draft one from the structure, or import them in bulk as described above.

Tip: Descriptions pay back later. The Schema Agent, the Query Agent, Pipelines and ontology tools all read them.