Skip to main content

Use conversational AI Private Preview

You can use conversational AI in Atlan to ask natural-language questions about your data, metadata, and business concepts.
It helps you explore assets, understand their relationships, and take guided actions without writing SQL or complex filters.

info

Make sure conversational AI is enabled for your tenant before you start. For setup, see Set up conversational AI.

Asset discovery prompts​

Use these prompts to discover assets, apply filters, and quickly find the most relevant and trusted objects in your catalog.

  • General exploration prompts: Use this to learn how conversational AI can support your day-to-day data discovery work.

    Example:

    How can you help me explore my data?

  • Trusted asset discovery prompts: Use this to quickly locate verified or certified assets that are safe to use.

    Example:

    Show me all verified tables.

  • Tag and custom metadata filter prompts: Use this to narrow down assets based on tags or custom metadata fields that matter to your teams.

    Examples:

    Can you give me a list of assets which have a tag of R&D?
    Share a list of dashboards with the custom metadata field Cost Attribution Centre set to COGS.

  • Owner and status filter prompts: Use this to find assets owned by a specific person and in a particular workflow state.

    Example:

    Can you list all the tables owned by shahrukh.khan which are in draft state?

  • Usage-based asset prompts: Use this to identify the most used dashboards so that you can focus on high-impact assets.

    Example:

    Give me a list of top 10 most used dashboards in our organization.

Definition and glossary prompts​

Use these prompts to understand how business terms, glossary entries, and data products are defined in Atlan.

  • Glossary topic prompts: Use this to understand what a glossary or subject area covers.

    Example:

    What is the glossary called Supply chain about?

  • Term definition and linked asset prompts: Use this to review how a specific term is defined, whether it's verified, and which assets are linked to it.

    Example:

    How is the term ARR defined in our organization? Is it verified? Can you list all the assets linked to this term?

  • Category term prompts: Use this to see how many terms exist in a category and what they represent.

    Example:

    What is inventory management? How many terms are there in this category?

  • Data product definition prompts: Use this to understand what a data product represents and who is responsible for it.

    Example:

    What is the Finance data product about? Who is the owner?

These prompts help you maintain consistent business language and connect business concepts to the right assets.

Asset comparison prompts​

Use these prompts to compare similar assets so that you can decide which one to use for a specific analysis or report.

  • Metric comparison prompts: Use this to understand differences between similar metrics and decide which one to use.

    Example:

    What is the difference between NPS in Customer Satisfaction and NPS in Customer Service KPIs? Which one should I use for quarterly reporting?

  • Domain asset group comparison prompts: Use this to compare related asset groups and see how they differ, including certification status.

    Example:

    Tell me the difference between Finance and Financial Products assets along with their certification status.

  • Table comparison prompts: Use this to compare several related tables and understand who owns each one.

    Example:

    Can you compare the tables sales_daily_report, sales_active_report, and sales_revenue_report? Who are the owners of each?

  • Aggregate table comparison prompts: Use this to compare aggregate tables and understand differences in purpose, freshness, and downstream usage.

    Example:

    Compare these two assets: SALES_DAILY_AGG and SALES_MONTHLY_AGG and highlight differences in purpose, freshness, and downstream usage.

These prompts help you choose the right asset for your use case and avoid duplicate or conflicting metrics.

Lineage analysis prompts​

Use these prompts to understand upstream and downstream lineage so that you can assess the impact of changes.

  • Downstream impact prompts: Use this to understand how changes to a table might affect downstream tables and dashboards.

    Example:

    What is the downstream impact of the table customer_order? If I delete it, will it have an impact on the table sales_order?

  • End-to-end lineage summary prompts: Use this to get a high-level summary of both upstream and downstream assets related to a key table.

    Example:

    Summarize the lineage of the table Opportunity. I want information on both upstream and downstream assets that are related to it.

  • Downstream dashboard and report prompts: Use this to find which dashboards or reports depend on a specific table.

    Example:

    Show me the assets downstream of the table CUSTOMER_TRANSACTIONS and summarize the types of dashboards that depend on it.

These prompts help you evaluate change impact and communicate dependencies to downstream consumers.

Atlan usage prompts​

Use these prompts to get step-by-step guidance on how to perform common tasks in Atlan.

  • Data product creation prompts: Use this to get a guided workflow for creating a data product and understanding what it represents.

    Example:

    Can you share a step-by-step process to create a data product in Atlan and explain what a data product is?

  • Column-level lineage setup prompts: Use this to learn how to configure and validate column-level lineage in Atlan.

    Example:

    How do I set up column-level lineage in Atlan?

  • Connection creation prompts: Use this to get step-by-step instructions for adding a new data connection.

    Example:

    How do I create a new connection to Snowflake in Atlan?

These prompts help you turn questions about how to use Atlan into clear, guided workflows.