Atlan MCP tools
The Atlan MCP server exposes 39 tools across nine categories: search and discovery, lineage and exploration, asset metadata, custom metadata, asset lifecycle, business glossary, data domains, data quality, and knowledge and artifacts. Each tool has a defined access level—Read (query only), Write (modifies metadata), or Admin (requires admin permissions). Filter by category or access level to find the tool you need.
Tools that change anything in Atlan propose first: they return a preview of the change and wait for your approval before writing. Some tools are gated per tenant, so your client may list fewer than the full set.
The same search agent that powers Conversational AI — search across the entire Atlan catalog using natural language with the same accuracy and intent-aware ranking in both places.
Hand an open-ended question to Atlan AI, which plans the work, queries the context layer, iterates, and returns one grounded conversational answer — complementary to semantic_search for questions you cannot break down yourself. It reasons before answering, so a single call typically takes 30–120 seconds.
Find assets using structured filters — type, certification status, tags, domains, glossary terms, and dates — with exclusions, OR-style matching, sorting, and aggregations. Returns 20 results per page by default, up to 100 per call, or just the total match count.
Retrieve complete details for one or more assets by GUID in a single call — metadata, owners, tags, custom attributes, and related entities such as columns, glossary terms, and README. Takes GUIDs from a prior search result, not names or qualified names, and returns catalog metadata only — use read_artifact for the contents of a file.
Search Atlan's product documentation and get back a generated answer with links to the source pages — for how-to and capability questions about Atlan itself, not for finding data assets.
Resolve tenant-level definitions — Atlan tags and custom metadata sets — by name or keyword, returning their exact names and GUIDs before another call uses them. Use the asset search tools for glossaries, terms, domains, and data products, get_users for people, and get_groups for groups.
List workspace groups and, optionally, their members, so AI agents can resolve group names to actual identities. Filter by name substring or exact group GUID.
Find workspace users by name, email, or username, or fetch one user by GUID — for owner and steward assignment, directory lookups, and resolving an owner GUID back to a person.
List the valid attributes and relationships for an asset type so a follow-up get_assets call can request exactly the type-specific fields it needs.
Trace upstream or downstream lineage for an asset to understand dependencies, data flows, and impact across your environment. Returns immediate neighbors by default; traverse deeper hops on request.
Run read-only SQL against a connected warehouse such as Snowflake, Redshift, BigQuery, or Databricks to retrieve real rows for analysis and validation. Statements that modify data are rejected, up to 100 rows are returned, and the caller needs query permission on the connection.
Modify asset metadata such as descriptions, certification status, and README content.
Set or update custom metadata attribute values on one or more assets. Partial by default, so unspecified attributes keep their values; full replacement clears them and applies to a single asset at a time.
Apply one or more classification tags to one or more assets. Propagation to downstream lineage is on by default and can be restricted to lineage only, hierarchy only, or turned off.
Remove a specific classification tag from one or more assets.
Add or remove informational, warning, or issue announcements on assets.
Define one or more custom metadata sets with typed attributes — string, number, boolean, date, enum, users, groups, URL, and more. Defines the schema only; use update_custom_metadata to set values on assets. Requires admin permissions in Atlan.
Add new typed attributes to an existing custom metadata set. Requires admin permissions in Atlan.
Archive specific attributes on a custom metadata set. Archived attributes are soft-deleted and their values are cleared from every asset. Requires admin permissions in Atlan.
Clear all custom metadata values from an asset for a specific metadata set.
Permanently delete an entire custom metadata set and all its attribute definitions. Requires admin permissions in Atlan.
Archive, restore, or purge assets to manage the end-of-life state of deprecated or replaced data assets. Purging is permanent and cannot be undone.
Define new business glossaries with metadata and descriptions so teams can organize terms by domain or function.
Add categories and subcategories inside glossaries so business terms are grouped into navigable hierarchies.
Create individual business terms with names, definitions, and certificate status.
Define data domains and subdomains to organize assets and data products by business area and ownership.
Create data products linked to assets within a domain so teams can publish and discover business-ready datasets.
Define rules on critical tables and views to enforce baseline data quality — completeness, uniqueness, validity, freshness, volume, reconciliation, and custom SQL checks.
Adjust thresholds, priorities, and conditions on existing rules so quality checks stay aligned with evolving data patterns.
Configure cron-based schedules for rules on tables and views so checks run automatically alongside ETL workflows.
Remove deprecated or noisy rules so your data quality rule set remains focused and maintainable.
Browse and read Atlan's artifact catalog — skills, knowledge files, and agents. Called with no arguments it lists everything with names, descriptions, and GUIDs; called with a GUID or name it returns the artifact as markdown with its file content inlined. Artifacts are stored separately from data assets, so the asset search tools do not surface them.
Search uploaded knowledge files semantically and get back focused passages with citations — file name, folder, page number, and document title. The primary way to pull business context out of documents. Currently in early preview and may not be enabled on every tenant.
List knowledge folders, or the files inside one folder, to see what context has already been uploaded before adding more. Currently in early preview and may not be enabled on every tenant.
Read the full text of one knowledge file — for summarizing, translating, or analyzing a whole document, or when the file has not been indexed for search yet. Currently in early preview and may not be enabled on every tenant.
Create a folder to organize uploaded documents such as PDFs, Markdown, and YAML that agents read for business context. Idempotent — returns the existing folder if it already exists. Currently in early preview and may not be enabled on every tenant.
Upload a document as a knowledge file and index it in the same call, so it is immediately searchable. Handles text formats directly and binary files such as PDF and DOCX up to 10 MB; identical content is deduplicated. Currently in early preview and may not be enabled on every tenant.
Get a presigned upload URL for large or binary knowledge files, to be paired with register_knowledge_file. Use save_knowledge_content instead for text or files under 5 MB. Currently in early preview and may not be enabled on every tenant.
Register a file in the catalog after a presigned upload completes, making it listable and readable. Currently in early preview and may not be enabled on every tenant.