AI Agents

Overview

The AI Tag Agent helps you keep metadata consistent and organized by automatically tagging tables, dashboards, and columns. Using natural language, administrators can define tagging rules that run daily to ensure metadata stays accurate and up to date.

Key Features

• Create tagging rules using natural language

• Preview tag assignments before saving

• Automatically execute rules daily

• Apply tags in bulk to tables, dashboards, and columns

• Validate and refine results using an interactive preview table

Getting Started

Prerequisites

• Admin access

• AI features enabled in the Admin Panel

Basic Workflow

1. Enable AI features in the Admin Panel.

2. Go to the AI Agents section in the Ask AI page.

3. Create a new agent by writing tagging instructions in natural language.

4. Review the preview to confirm the agent’s interpretation.

5. Save the automation to schedule daily execution.

Example Rules

You can define rules based on different criteria, such as names, dates, lineage, popularity, or existing tags:

• By name: Add a PII tag to all columns named "customer_email"

• By date: Tag tables created in the last 30 days with "new_data"

• By lineage: Tag tables with no downstream dependencies as "endpoint"

• By popularity: Tag dashboards with popularity > 80 as "gold"

• By existing tags: Add "sensitive" tag to all tables and columns already tagged with "PII"

How It Works

When you create a rule, the AI Tag Agent will:

1. Identify all items matching your criteria.

2. Show a preview of the proposed tag assignments.

3. Allow you to confirm or adjust before saving.

4. Run the approved rules automatically every day.