Atlan

Atlan MCP Server

MCP server for interacting with Atlan services including asset search, updates, and lineage traversal for comprehensive data governance and discovery.

What is an MCP Server?

MCP Info

Image Building Info

Available Tools (14)

Tools Details

Tool: create_data_products

Create Data Products in Atlan.

IMPORTANT BUSINESS RULES & CONSTRAINTS:

• Before creating a product, you may want to search for existing domains to get the qualified_name for the domain relationship
• Product names must be unique within the same domain
• At least one asset GUID must be provided for each product

Tool: create_domains

Create Data Domains or Sub Domains in Atlan.

IMPORTANT BUSINESS RULES & CONSTRAINTS:

• Before creating a domain/subdomain, you may want to search for existing domains to avoid duplicates or to get the qualified_name for parent relationships
• Domain names must be unique at the top level
• Subdomain names must be unique within the same parent domain

Tool: create_dq_rules_tool

Create one or multiple data quality rules in Atlan.

Supports all rule types: column-level, table-level, and custom SQL rules. Rules can be created individually or in bulk for efficient setup.

• rule_type (str): Type of rule (see Supported Rule Types) [REQUIRED]
• asset_qualified_name (str): Qualified name of the asset (Table, View, MaterialisedView, or SnowflakeDynamicTable) [REQUIRED]
• asset_type (str): Type of asset - "Table" | "View" | "MaterialisedView" | "SnowflakeDynamicTable" [OPTIONAL, default: "Table"]
• threshold_value (int/float): Threshold value for comparison [REQUIRED]
• column_qualified_name (str): Column qualified name [REQUIRED for column-level rules, NOT for Row Count/Custom SQL]
• threshold_compare_operator (str): Comparison operator (EQUAL, GREATER_THAN, etc.) [OPTIONAL, default varies by rule]
• threshold_unit (str): Time unit for Freshness rules (DAYS, HOURS, MINUTES) [REQUIRED for Freshness, N/A for others]
• alert_priority (str): Alert priority level (LOW, NORMAL, URGENT) [OPTIONAL, default: NORMAL]
• row_scope_filtering_enabled (bool): Enable row-level filtering [OPTIONAL]
• rule_conditions (List[Dict]): Conditions for String Length/Regex/Valid Values [REQUIRED for conditional rules]
• custom_sql (str): SQL query [REQUIRED for Custom SQL rules]
• rule_name (str): Name for the rule [REQUIRED for Custom SQL rules]
• dimension (str): DQ dimension [REQUIRED for Custom SQL rules]
• description (str): Rule description [OPTIONAL]

Tool: create_glossaries

Create one or multiple AtlasGlossary assets in Atlan.

IMPORTANT BUSINESS RULES & CONSTRAINTS:

• Check for duplicate names within the same request and ask user to choose different names
• Do NOT use search tool before creating glossaries - Atlan will handle existence validation
• If user gives ambiguous instructions, ask clarifying questions

• name (str): Name of the glossary (required)
• user_description (str, optional): Detailed description of the glossary proposed by the user
• certificate_status (str, optional): Certification status ("VERIFIED", "DRAFT", or "DEPRECATED")

Tool: create_glossary_categories

Create one or multiple AtlasGlossaryCategory assets in Atlan.

IMPORTANT BUSINESS RULES & CONSTRAINTS:

• There cannot be two categories with the same name under the same glossary (at the same level)
• Under a parent category, there cannot be subcategories with the same name (at the same level)
• Categories with the same name can exist under different glossaries (this is allowed)
• Cross-level naming is allowed: category "a" can have subcategory "b", and category "b" can have subcategory "a"
• Example allowed structure: Glossary "bui" → category "a" → subcategory "b" AND category "b" → subcategory "a"
• Always check for duplicate names at the same level and ask user to choose different names
• Before creating a category, perform a single search to check if the glossary or categories with the same name already exist. Skip this step if you already have the required GUIDs.
• Example call for searching glossary and categories before category creation(Query - create categories Locations and Characters under Marvel Cinematic Universe (MCU) glossary): { "limit": 10, "conditions": { "type_name": ["AtlasGlossary", "AtlasGlossaryCategory"], "name": ["Marvel Cinematic Universe (MCU)", "Characters", "Locations"] } }
• If user gives ambiguous instructions, ask clarifying questions

• name (str): Name of the category (required)
• glossary_guid (str): GUID of the glossary this category belongs to (required)
• user_description (str, optional): Detailed description of the category proposed by the user
• certificate_status (str, optional): Certification status ("VERIFIED", "DRAFT", or "DEPRECATED")
• parent_category_guid (str, optional): GUID of the parent category if this is a subcategory

Tool: create_glossary_terms

Create one or multiple AtlasGlossaryTerm assets in Atlan.

IMPORTANT BUSINESS RULES & CONSTRAINTS:

• Within a glossary, a term (single GUID) can be associated with many categories
• Two terms with the same name CANNOT exist within the same glossary (regardless of categories)
• A term is always anchored to a glossary and may also be associated with one or more categories inside the same glossary
• Before creating a term, perform a single search to check if the glossary, categories, or term with the same name already exist. Search for all relevant glossaries, categories, and terms in one call. Skip this step if you already have the required GUIDs.
• Example call for searching glossary categories and terms before term creation(Query - create a term fighterz under category Characters and Locations under Marvel Cinematic Universe (MCU) glossary): { "limit": 10, "conditions": { "type_name": ["AtlasGlossary", "AtlasGlossaryCategory","AtlasGlossaryTerm"], "name": ["Marvel Cinematic Universe (MCU)", "Characters", "Locations","fighterz"] } }

• name (str): Name of the term (required)
• glossary_guid (str): GUID of the glossary this term belongs to (required)
• user_description (str, optional): Detailed description of the term proposed by the user
• certificate_status (str, optional): Certification status ("VERIFIED", "DRAFT", or "DEPRECATED")
• category_guids (List[str], optional): List of category GUIDs this term belongs to.

Tool: delete_dq_rules_tool

Delete one or multiple data quality rules in Atlan.

Tool: get_assets_by_dsl_tool

Execute the search with the given query dsl_query : Union[str, Dict[str, Any]] (required): The DSL query used to search the index.

Example: dsl_query = '''{ "query": { "function_score": { "boost_mode": "sum", "functions": [ {"filter": {"match": {"starredBy": "john.doe"}}, "weight": 10}, {"filter": {"match": {"certificateStatus": "VERIFIED"}}, "weight": 15}, {"filter": {"match": {"certificateStatus": "DRAFT"}}, "weight": 10}, {"filter": {"bool": {"must_not": [{"exists": {"field": "certificateStatus"}}]}}, "weight": 8}, {"filter": {"bool": {"must_not": [{"terms": {"__typeName.keyword": ["Process", "DbtProcess"]}}]}}, "weight": 20} ], "query": { "bool": { "filter": [ { "bool": { "minimum_should_match": 1, "must": [ {"bool": {"should": [{"terms": {"certificateStatus": ["VERIFIED"]}}]}}, {"term": {"__state": "ACTIVE"}} ], "must_not": [ {"term": {"isPartial": "true"}}, {"terms": {"__typeName.keyword": ["Procedure", "DbtColumnProcess", "BIProcess", "MatillionComponent", "SnowflakeTag", "DbtTag", "BigqueryTag", "AIApplication", "AIModel"]}}, {"terms": {"__typeName.keyword": ["MCIncident", "AnomaloCheck"]}} ], "should": [ {"terms": {"__typeName.keyword": ["Query", "Collection", "AtlasGlossary", "AtlasGlossaryCategory", "AtlasGlossaryTerm", "Connection", "File"]}}, ] } } ] }, "score_mode": "sum" }, "score_mode": "sum" } }, "post_filter": { "bool": { "filter": [ { "bool": { "must": [{"terms": {"__typeName.keyword": ["Table", "Column"]}}], "must_not": [{"exists": {"field": "termType"}}] } } ] }, "sort": [ {"_score": {"order": "desc"}}, {"popularityScore": {"order": "desc"}}, {"starredCount": {"order": "desc"}}, {"name.keyword": {"order": "asc"}} ], "track_total_hits": true, "size": 10, "include_meta": false }''' response = get_assets_by_dsl(dsl_query)

Tool: query_asset_tool

Execute a SQL query on a table/view asset.

This tool enables querying table/view assets on the source similar to what's available in the insights table. It uses the Atlan query capabilities to execute SQL against connected data sources.

CRITICAL: Use READ-ONLY queries to retrieve data. Write and modify queries are not supported by this tool.

Tool: schedule_dq_rules_tool

Schedule data quality rule execution for one or multiple assets.

• asset_type (str): "Table", "View", "MaterialisedView", or "SnowflakeDynamicTable"
• asset_name (str): Name of the asset
• asset_qualified_name (str): Qualified name of the asset
• schedule_crontab (str): Cron expression (5 fields: min hour day month weekday)
• schedule_time_zone (str): Timezone (e.g., "UTC", "America/New_York")

Tool: search_assets_tool

Advanced asset search using FluentSearch with flexible conditions.

Tool: traverse_lineage_tool

Traverse asset lineage in specified direction.

By default, essential attributes are included in results. Additional attributes can be specified via include_attributes parameter for richer lineage information.

Tool: update_assets_tool

Update one or multiple assets with different values for attributes or term operations.

Tool: update_dq_rules_tool

Update existing data quality rules in Atlan.

• qualified_name: Rule's qualified name
• rule_type: Rule type (e.g., "Null Count", "Row Count", "Custom SQL")
• asset_qualified_name: Table/view qualified name

Use this MCP Server

{
  "mcpServers": {
    "atlan": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ATLAN_API_KEY",
        "-e",
        "ATLAN_BASE_URL",
        "mcp/atlan"
      ],
      "env": {
        "ATLAN_API_KEY": "<YOUR_ATLAN_API_KEY>",
        "ATLAN_BASE_URL": "https://your-instance.atlan.com"
      }
    }
  }
}

Why is it safer to run MCP Servers with Docker?