--- title: "The Atlan MCP Cookbook | Recipes for AI that understands your data" url: "https://atlan.com/mcp-cookbook/" description: "Real, tested recipes for the Atlan MCP. Enrich, govern, and put your metadata to work as context your AI can actually use - from inside any AI tool." keywords: "atlan mcp, mcp cookbook, ai context, model context protocol, metadata for ai, data catalog mcp, ai data governance" --- > Atlan is hosting Context Conference, bringing together the leaders and builders at the frontier of giving AI the context it needs to understand their business. It runs online on October 28, 2026, from 11:00 AM to 2:00 PM ET. Atlan co-founder Prukalpa Sankar opens and closes the day. Leaders from AstraZeneca, BNY and Verizon share why they invest in context and what they get from it. Registrants get early access to The AI Context Gap, a new study from MIT Technology Review Insights. Register: https://atlan.com/context-conference/ **The Atlan MCP Cookbook: see what your AI can do with your context.** Real, tested recipes for the Atlan MCP. Your company's context, live in whatever tool you already build in. The page lists 33 recipes, filterable by use case, persona (Data/Business User, Data/BI Analyst, AI Builder, Data Engineer, Data Governance Lead) and Atlan action type (Look up or Make changes). [Set up Atlan MCP](https://docs.atlan.com/product/capabilities/atlan-ai/how-tos/remote-mcp-overview) ## How the recipes work 1. **Ask in your AI tool:** type the prompt and let it work. 2. **Automate it:** run it on a schedule or from a pipeline. 3. **Chain the steps:** search, trace lineage, then update, end to end. Each recipe line below reads: name (link to the recipe page), action type, effort, personas it serves, then what it does. ## Find and understand data (13 recipes) - [Route data requests to right assets](https://atlan.com/mcp-cookbook/route-data-requests-to-right-assets/) - Look up / Quick / Data/Business User, Data/BI Analyst, Data Governance Lead. Surface certified, well-documented assets in response to a business request. - [Understand how a metric is calculated](https://atlan.com/mcp-cookbook/understand-how-a-metric-is-calculated/) - Look up / Quick / Data/Business User, Data/BI Analyst, AI Builder. Get the governed definition of a metric - and where it's documented - without asking a data engineer. - [Reuse before you rebuild](https://atlan.com/mcp-cookbook/reuse-before-you-rebuild/) - Look up / Quick / AI Builder, Data Engineer. Before you write SQL to derive a metric, check whether it already exists - and reuse the governed one. - [Get build-ready for a new dashboard](https://atlan.com/mcp-cookbook/get-build-ready-for-a-new-dashboard/) - Look up / Moderate / Data/BI Analyst, AI Builder, Data Engineer. Before you build a dashboard: the trusted source, the confirmed definition, and how the data is really used - as the build recipe. - [Build a new table or view on trusted sources](https://atlan.com/mcp-cookbook/build-a-new-table-or-view-on-trusted-sources/) - Look up / Moderate / AI Builder, Data Engineer. Before you write the DDL, use Atlan to pick the right sources, confirm the grain and definitions, and avoid rebuilding something that already exists. - [Talk to your data](https://atlan.com/mcp-cookbook/talk-to-your-data/) - Look up / Quick / Data/Business User, Data/BI Analyst, AI Builder. Pair Atlan with your warehouse or SQL-tool MCP: Atlan finds the right certified table and hands over the governed definition, so you query the source correctly - not just quickly. - [Write SQL that runs on the first try](https://atlan.com/mcp-cookbook/explore-schema-before-writing-sql/) - Look up / Quick / Data/BI Analyst, AI Builder, Data Engineer. Pull real column names and types from the source schema first, so the SQL you write runs on the first try instead of guessing at a field name that doesn't exist. - [Decode internal jargon without asking around](https://atlan.com/mcp-cookbook/decode-internal-jargon-against-the-glossary/) - Look up / Quick / Data/Business User, Data/BI Analyst, AI Builder. Check whether an internal acronym or jargon term already has a governed definition before you interrupt five people on Slack to ask. - [Know what your domain owns, and whether it's live](https://atlan.com/mcp-cookbook/navigate-data-products-by-domain/) - Look up / Moderate / Data/Business User, Data/BI Analyst, Data Governance Lead. Ask what data products exist under a business domain and what actually backs one of them, instead of hunting table by table. - [Get the answer, not just the table](https://atlan.com/mcp-cookbook/get-the-answer-not-just-the-table/) - Look up / Moderate / Data/Business User, Data/BI Analyst, AI Builder. Every other lookup hands you where to find the data. This one starts from the governed definition of a metric, follows it to the certified asset that carries it, and gives you the answer itself - calculated the way your business defines it, using Atlan alone. - [Profile a field's shape before you build on it](https://atlan.com/mcp-cookbook/profile-a-field-before-you-build-on-it/) - Look up / Quick / Data/BI Analyst, AI Builder, Data Engineer. Get a quick profile of a field, its type, null rate, distinct count, and range, so you know what you're working with before you aggregate or join on it. - [Find the source you can actually trust](https://atlan.com/mcp-cookbook/find-the-trustworthy-replacement-source/) - Look up / Moderate / Data/Business User, Data/BI Analyst, AI Builder, Data Engineer, Data Governance Lead. When the table you've always been pointed to stops reconciling, find the actual certified source instead of guessing at a second bad table. - [Ground AI-written code in your real schema](https://atlan.com/mcp-cookbook/ground-code-in-real-schema/) - Look up / Quick / AI Builder, Data Engineer. Pull exact column names, types, and certification status for the tables your fix touches, before your coding agent writes a line of SQL. ## Change and fix data safely (4 recipes) - [See what a SQL change will break before you merge](https://atlan.com/mcp-cookbook/see-what-a-sql-change-will-break-before-you-merge/) - Look up / Quick / AI Builder, Data Engineer. Find every downstream consumer of a changed column or table before merging. - [Warn downstream owners before a change lands](https://atlan.com/mcp-cookbook/warn-downstream-owners-before-a-change-lands/) - Make changes / Moderate / AI Builder, Data Engineer. Find every downstream consumer of a changing asset and warn their owners first. - [Triage a data-quality failure with full context](https://atlan.com/mcp-cookbook/triage-a-data-quality-failure-with-full-context/) - Make changes / Moderate / Data Engineer. Enrich quality failures with lineage, ownership, and downstream impact for faster triage. - [Trace a broken dashboard to its root cause](https://atlan.com/mcp-cookbook/trace-a-broken-dashboard-to-its-root-cause/) - Look up / Moderate / Data/BI Analyst, Data Engineer. Trace lineage upstream from a broken dashboard to find the source of wrong numbers. ## Retire and migrate (3 recipes) - [Safely deprecate unused assets](https://atlan.com/mcp-cookbook/safely-deprecate-unused-assets/) - Make changes / Moderate / Data Engineer. Verify no active consumers exist, notify owners, then mark assets as deprecated. - [Triage assets for migration](https://atlan.com/mcp-cookbook/triage-assets-for-migration/) - Make changes / Involved / Data Engineer. Classify every asset by consumer count, ownership, and certification before migrating. - [See exactly what a migration would leave behind](https://atlan.com/mcp-cookbook/compare-two-sources-before-a-migration-decision/) - Look up / Moderate / Data Engineer. Line up two connections, databases, or schemas and see exactly what matches, what's missing, and what only exists on one side, before you commit to a migration or consolidation plan. ## Make data understandable (6 recipes) - [Enrich assets with your context](https://atlan.com/mcp-cookbook/enrich-assets-with-your-context/) - Make changes / Quick / Data/BI Analyst, Data Engineer, Data Governance Lead. Bring context that lives outside Atlan, like a spreadsheet or a Confluence page, onto your assets as owners, tags, and additional metadata. - [Give your dashboards context](https://atlan.com/mcp-cookbook/give-your-dashboards-context/) - Make changes / Quick / Data Governance Lead, Data/BI Analyst. Link BI dashboards to business terms and owners so they show up in search. - [Propagate context through lineage](https://atlan.com/mcp-cookbook/propagate-context-through-lineage/) - Make changes / Moderate / Data/BI Analyst, Data Engineer, Data Governance Lead. Push descriptions and tags from a source table to every downstream asset in one pass. - [Grow your business glossary](https://atlan.com/mcp-cookbook/grow-your-business-glossary/) - Make changes / Moderate / Data Governance Lead, Data/BI Analyst. Bootstrap glossary terms from how your assets already use business concepts. - [Document SQL with business context](https://atlan.com/mcp-cookbook/document-sql-with-business-context/) - Make changes / Moderate / AI Builder, Data Engineer. Generate business-readable descriptions for SQL models from the context already in Atlan. - [Give one concept one trusted definition](https://atlan.com/mcp-cookbook/reconcile-terminology-across-systems/) - Make changes / Moderate / Data Governance Lead, Data/BI Analyst. Confirm whether two differently-named fields or terms mean the same thing, then link them under one governed definition so the next person searching either name lands in the same place. ## Govern and keep trusted (7 recipes) - [Keep asset metadata fresh](https://atlan.com/mcp-cookbook/keep-asset-metadata-fresh/) - Make changes / Moderate / Data Governance Lead. Catch stale certifications, missing owners, and outdated descriptions early. - [Sync metadata across environments](https://atlan.com/mcp-cookbook/sync-metadata-across-environments/) - Make changes / Moderate / Data Governance Lead, Data Engineer. Keep descriptions, tags, and custom metadata aligned across dev, staging, and prod. - [Propagate PII tags to parent assets](https://atlan.com/mcp-cookbook/propagate-pii-tags-to-parent-assets/) - Make changes / Moderate / Data Engineer, Data Governance Lead. Roll column-level PII classification up to parent tables and schemas so policies fire. - [Score assets for AI readiness](https://atlan.com/mcp-cookbook/score-assets-for-ai-readiness/) - Make changes / Moderate / AI Builder, Data Governance Lead. Rate assets on documentation, ownership, and quality before they enter AI pipelines. - [Gamify your context](https://atlan.com/mcp-cookbook/gamify-your-context/) - Look up / Quick / Data Governance Lead. Generate per-domain enrichment scorecards to motivate metadata contributions. - [Check the master list before you add a new value](https://atlan.com/mcp-cookbook/check-value-against-master-list/) - Look up / Quick / Data/Business User, Data/BI Analyst, Data Governance Lead. Before adding a new value to a field, pull the existing values so a new entry doesn't duplicate one under a different name and fragment the master list. - [Reach the real owner, not the wrong person](https://atlan.com/mcp-cookbook/find-the-right-owner-to-contact/) - Look up / Quick / Data/Business User, Data Engineer, Data Governance Lead. Trace a failing number back to the asset behind it, then find out who owns that dataset, so the problem reaches the team accountable for it instead of a general channel. ## Ready to cook? Point your AI at the Atlan MCP and run your first recipe: [Set up Atlan MCP](https://docs.atlan.com/product/capabilities/atlan-ai/how-tos/remote-mcp-overview) / [Read the docs](https://docs.atlan.com/product/capabilities/atlan-ai/how-tos/atlan-mcp-overview)