Model context protocol (MCP)
The Model Context Protocol (MCP) is an open standard that enables AI agents to access contextual metadata from external systems. It provides a consistent way for large language models and automation frameworks to retrieve the context they need to generate accurate and reliable results.
Atlan MCP is based on this standard and provides a reference implementation through the Atlan MCP server. The server acts as a secure bridge between Atlan’s metadata platform and AI tools such as Claude, Cursor, Windsurf, and Microsoft Copilot Studio. With Atlan MCP, you can search and discover assets, explore lineage, update metadata, create glossaries, and more, all using real-time context from Atlan.
Atlan MCP tools
The Atlan MCP server provides a set of tools that enable AI agents to work directly with Atlan metadata. These tools supply real-time context to AI environments, making it easier to search, explore, and update metadata without leaving your workflow.
The latest version of the server includes all of the tools listed below. If you don't see a tool in your environment, update to the latest Atlan MCP server version.
Search assets
Find assets in Atlan using flexible filters for name, type, tags, and domains so AI agents surface the most relevant context.
Query by DSL
Retrieve specific assets using Atlan's DSL query language to run precise lookups beyond basic search filters.
Query assets
Run read-only SQL queries against tables and views so AI agents can preview and analyze data using existing Atlan connections.
Explore lineage
Trace upstream or downstream lineage for an asset to understand dependencies, data flows, and impact across your environment.
Update assets
Modify metadata such as descriptions, certification status, and README content so AI workflows keep asset context up to date.
Glossary
Explore existing glossaries, terms, and relationships so AI agents can use consistent business definitions across tools.
Create glossaries
Define new business glossaries with metadata and descriptions so teams can organize terms by domain or function.
Create glossary categories
Add categories and subcategories inside glossaries so business terms are grouped into clear, navigable hierarchies.
Create glossary terms
Create individual business terms with names, definitions, and certificate status so AI agents can align prompts with shared language.
Data domains and products
Define data domains, subdomains, and data products linked to assets so AI agents can reason about business-ready datasets and contracts.
Create data quality rules
Define column, table, and SQL-based rules on critical assets so AI workflows can enforce baseline data quality checks.
Update data quality rules
Adjust thresholds, priorities, and conditions on existing rules so quality checks stay aligned with evolving data patterns.
Schedule data quality rules
Configure cron-based schedules for rules on tables and views so data quality checks run automatically alongside ETL workflows.
Delete data quality rules
Remove deprecated or noisy rules so your data quality rule set remains focused and maintainable.
Deployment options
You can connect to Atlan MCP in two ways:
Hosted, per-tenant MCP server managed by Atlan, with OAuth and API Key authentication. Ideal for production setups and available in private preview for eligible tenants.
Locally hosted MCP server you run with Docker or uv. Ideal for development, testing, and custom environments before moving to a managed option.