Govern catalog quality
Use these workflows to automate routine governance operations from your AI client—keeping asset metadata current, enriching quality alerts with lineage and ownership context, propagating classification tags through the asset hierarchy, and scoring assets for AI readiness. Governance tasks such as recertification, PII classification, and environment sync require regular attention but are time-consuming to perform manually across a large catalog. These workflows handle them through natural language prompts without requiring direct access to the Atlan UI.
Make sure Atlan MCP is configured before running these workflows. For setup, see Set up Atlan MCP.
Keep asset metadata fresh
Metadata drifts over time—certifications expire, teams change, and descriptions go stale. This workflow surfaces assets with outdated or missing metadata so you can fix them in bulk before they cause confusion downstream.
Find all tables in the analytics schema certified more than 6 months ago. Review their current descriptions and owners, and flag any that need recertification with a warning announcement.
Context-rich data quality alerts
A bare data quality alert—'null check failed on orders_fact'—tells you what broke but not who owns it, what depends on it, or how urgent it is. This workflow enriches the alert with lineage, ownership, and downstream impact so the right people can triage it immediately.
The orders_fact table failed a null check on order_id this morning. Find its owners and downstream dashboards, then add a warning announcement with the failure details so the team can triage.
Asset change alerts
Schema changes—a renamed column, a dropped field, a modified type—can cause downstream pipelines and dashboards to fail if the owners of those assets aren't notified in advance. This workflow traverses lineage to find every consumer of a changed asset and adds an announcement to each one before the change is applied.
The customer_id column in dim_customers is being renamed to cust_id next week. Find all downstream tables and dashboards that use it and add a warning announcement explaining the change and timeline.
Sync metadata across environments
Dev, staging, and production versions of the same asset often have mismatched descriptions, tags, and custom metadata. This workflow compares them and syncs the more complete version so your environments stay consistent.
Find the orders_fact table in both production and staging. Compare their descriptions, tags, and custom metadata, then sync the production metadata to staging where staging is missing fields.
Propagate PII tags to parent assets
When a column is tagged as PII, its parent table and schema must also be classified—otherwise access controls and governance policies that apply at the table or schema level won't fire. This workflow walks up the hierarchy and applies missing classification tags.
Find all columns tagged as PII in the customers database. For each one, apply the PII tag to its parent table and schema if those aren't already tagged.
Score assets for AI readiness
Before an asset can be safely used in an AI pipeline, it needs a description, a clear owner, a certification, and a PII classification. This workflow evaluates each asset on those criteria and writes a readiness score to a custom metadata field so teams know what's ready.
For all tables in the ml_features schema, check if they have descriptions, owners, certifications, and PII classification. Score each one for AI readiness and write the score to the AI Readiness custom metadata field.