Catalog
1. Overview
Purpose of the Catalog
The Catalog in Connecty AI provides a centralized view of your data assets. It acts as a structured directory where you can explore the contents of your connected databases — including tables, columns, schemas and metadata. Whether you’re a data analyst, engineer or business user, the Catalog helps you understand what data is available, how it's organized, and what it means — all without writing a single line of code.
When and why to use it
You’ll find the Catalog particularly useful when:
You’re exploring a new dataset and want to quickly understand its structure.
You need to locate a specific table or column to build queries or validate data definitions.
You're reviewing metadata, such as row counts, data types, or descriptions, to assess data quality.
You want to identify columns that may contain Personally Identifiable Information (PII), either manually marked or automatically suggested by the system, so you can handle sensitive data appropriately.
You’re collaborating across teams and need to align on the meaning and purpose of certain data elements.
Using the Catalog before diving into query building or analysis gives you clarity and saves time by removing guesswork.
Where to find it
You can access the Catalog by navigating to the Context Engine section in the left-hand menu. Once there, select Catalog from the list of available tools. This will open the Catalog interface, where you can switch between it tabs. Each tab presents a different level of detail, allowing you to explore from the top level (entire databases) down to specific columns.
Interactive Entity Inspector
Clicking on any item in the Catalog — whether it’s a database, schema, object or column — opens a detailed Entity Inspector. This inspector reveals additional metadata such as data definitions, sample values, statistics, DDL statements and system flags. This interactive layer helps you drill deeper into the structure and meaning of your data without needing to leave the Catalog view.
2. Catalog structure
2.1 Hierarchy and tabs connections
The Catalog is organized to reflect the logical structure of a database environment, with information grouped into four tabs: Database, Schema, Objects, and Columns. These tabs are interconnected and follow a top-down hierarchy, moving from general to specific.
The structure makes it easy to drill down from a high-level overview (what data sources are available?) to very detailed metadata (what does this specific column contain?). You can also click on any entity in these tabs to reveal deeper information.
2.2 Database tab
The Database tab is the entry point of the Catalog and represents the highest level of your data environment. It lists all available databases connected to your workspace and provides a quick overview of each one’s purpose and contents.
What you’ll see
For each database, the following details are shown:
Database Name – The name of the connected database (e.g.,
connecty-ai).Schema Count – How many schemas are inside that database.
Tables Count – Total number of tables (objects) found within all schemas.
Enriched Description – A short, editable summary explaining what kind of data the database holds.
How to use it
Use the Database tab when you want to:
Get a high-level understanding of available databases.
Quickly compare schema and table volumes across sources.
Orient yourself before exploring the data further.
Add or read a short description that helps your team understand the database's purpose.
Click for more
Clicking on a database opens a description panel, where you can:
View its enriched description
Edit the description using the pencil icon
Use the description to document the database’s role, content or any special considerations — helpful when working in teams or managing multiple data sources.
2.3 Schema tab
The Schema tab shows the second level in the Catalog hierarchy. A schema is a logical container within a database, grouping together related tables and other data objects.
What you’ll see
Each row in the Schema tab represents one schema and includes:
Schema Name – The technical name of the schema.
Database Name – The parent database the schema belongs to.
Tables Count – Number of tables (objects) stored within that schema.
Enriched Description – A human-readable explanation of the schema’s purpose and contents.
How to use it
Use the Schema tab when you want to:
Understand how the database is organized
See what schemas exist and how many tables each one contains
Read or write short descriptions to clarify what each schema holds
Locate the appropriate schema before diving into its tables or columns
Click for more
Clicking on a schema opens a description panel, where you can:
View its enriched description
Edit the description using the pencil icon
Think of the Schema tab as a zoomed-out view of your data landscape. Use it to decide which parts of the database you want to investigate further via the Objects and Columns tabs.
2.4 Objects tab
The Objects tab brings you to the table level — this is where actual data lives. In most cases, “objects” refer to tables, though other object types (like views) could be included depending on the system setup.
This tab helps you understand what datasets are available, how large they are, and what kind of information they contain.
What you’ll see
Each row in the Objects tab corresponds to one table and includes:
Name – The name of the table (e.g.,
customer,orders,lineitem)Type – The object type (e.g.,
table)State – Whether the object is currently available
Database – The database the table belongs to
Schema – The schema that contains the object
Enriched Description – A short explanation of what kind of data the table holds
Row Count – Total number of records
Size (MB) – Storage size of the table
Columns Count – How many columns the table contains
How to use it
Use the Objects tab when you want to:
Quickly find a table based on name or description
Identify large tables by row count or storage size
Understand the purpose of a table through its enriched description
Get a high-level sense of how detailed or complex a table is (via columns count)
Click for more
Clicking on an object opens the object (table) inspector, which includes:
Full DDL (Data Definition Language) statement — showing how the table is structured
Column list with names, data types, descriptions, and sample values
Basic stats such as row count, size, and timestamps
A field for editing the table’s description
2.5 Columns tab
The Columns tab provides the most granular view in the Catalog. It lists all individual columns across all tables and gives detailed metadata about each one, making it easy to understand field-level data even before querying anything. This tab is especially useful for data exploration, validation and governance.
What you’ll see
Each row in the Columns tab represents a single column and includes:
ID – Internal identifier for the column
Name – The technical name of the column (e.g.,
_fivetran_deleted)Data Type – The abstracted type (e.g., BOOL, STRING, TIMESTAMP)
Table – The table the column belongs to
Schema – The schema where the table resides
Database – The parent database
State – Whether the column is currently available
Enriched Description – A short explanation of what the column represents
PII – Indicates if the column has been marked as containing Personally Identifiable Information
AI Suggested PII – Indicates if the system suspects the column may contain PII
Deleted – Whether the column is marked as deleted in the source
How to use it
Use the Columns tab when you want to:
Look up a column by name across all tables
Understand column-level metadata (e.g., data types, purpose, privacy sensitivity)
Identify fields that may contain sensitive information
Validate whether a certain attribute exists in your dataset before writing queries
Explore enriched descriptions for better understanding of field roles
Click for more
Clicking on a column opens the column inspector, where you’ll find:
Full metadata including nullability, position, and data type
Column statistics (row count, null percentage, sample values)
Flags like system-defined, hidden, or partitioning column
Editable enriched description
PII-related settings (user-defined and AI-suggested)
Last updated