Collibra MCP Server (Local Version)

Overview

The Collibra Model Context Protocol (MCP) server enables AI/ML developers, business users, and LLMs to access and engage with governed metadata and business context information in the Collibra Platform. This integration simplifies the connection of Collibra with agents built on LLMs and other agents in complex workflows.

The MCP server provides a standardized access layer that addresses three key challenges: inconsistent or low-quality context retrieval for LLMs, fragmented integration workflows, and lack of standardized communication between Collibra and other information sources.

This open-source offering is available for direct download and installation by developers through the linked GitHub repository or via package managers like Homebrew.

Read Tools

1. discover_business_glossary | Searches for glossary definitions via Collibra AI Copilot.
   1. Note: this requires Collibra AI Copilot Public Preview to be enabled
2. discover_data_assets | Searches for data assets via Collibra AI Copilot. 
   1. Note: this requires Collibra AI Copilot Public Preview to be enabled
3. get_asset_details | Retrieves a comprehensive, pre-formatted summary of a specific data asset by its ID
4. get_business_term_data | Trace a business term back to its connected physical data assets
5. get_column_semantics | Retrieve data attributes, measures, and business assets connected to a column
6. get_lineage_downstream | Get downstream technical lineage (consumers) for a data entity
7. get_lineage_entity | Get metadata about a specific entity in the technical lineage graph
8. get_lineage_transformation | Get details and logic of a specific data transformation
9. get_lineage_upstream | Get upstream technical lineage (sources) for a data entity
10. get_measure_data | Trace a measure back to its underlying physical columns and tables
11. get_table_semantics | Retrieve the semantic layer for a table: columns, data attributes, and connected measures
12. list_asset_types | Lists available asset types in the system. 
13. list_data_contract | Retrieves the details of existing data contracts. Use this for general queries to view data contracts or get contract details, and always return both their IDs and names in the response.
    1. Note: requires Data Contracts Private Preview to be enabled
14. prepare_add_business_term | Validate business term data, resolve domains, check for duplicates, and hydrate attribute schemas. Always call this before add_business_term.
15. prepare_create_asset | Resolve asset type, domain, hydrate full attribute schema, and check duplicates. Always call this before create_asset.
16. pull_data_contract_manifest | Downloads the manifest file for the currently active version of a specific data contract. 
    1. Note: requires Data Contracts Private Preview to be enabled
17. search_asset_keyword | Searches for data assets, business terms, communities, or other resources. This tool allows filtering by name, type, status, community or domain.
18. search_data_class | Retrieves a list of available data classes with filtering capabilities. 
19. search_data_classification_match | Finds associations between data classes and assets.. Supports filtering by asset IDs, classification IDs, statuses, and asset types.
20. search_lineage_entities | Search for entities in the technical lineage graph
21. search_lineage_transformations | Search for transformations in the technical lineage graph

Write Tools

1. add_business_term | Create a business term asset with definition and optional attributes in Collibra.
2. add_data_classification_match | Assigns a data classification to a specific data asset. This tool allows for associating a data class with an asset. 
3. create_asset | Create a new data asset with optional attributes in Collibra.
4. push_data_contract_manifest | Upload a new version of a data contract manifest to Collibra. The manifestID and version are automatically parsed from the manifest content if it adheres to the Open Data Contract Standard.
   1. Note: requires Data Contracts Private Preview to be enabled
5. remove_data_classification_match | Removes a data classification from a specific data asset. This tool allows for the removal of a data class from an asset. 
6. edit_asset | Edit an existing asset via a list of typed operations:
   1. update_attribute, add_attribute, remove_attribute – change, append, or clear an attribute value (e.g. Definition, Note)
   2. update_property – rename the asset (name), change its displayName, or change its statusId (status name or UUID accepted)
   3. add_relation, remove_relation – link or unlink the asset to another asset by relation role (e.g. is synonym of)
   4. add_tag – append a free-text tag without replacing existing tags

Documentation

For installation instructions and detailed configuration steps, please refer to the README in the GitHub repository.

Notes

• For our first release, the binaries will not be signed and therefore the binary files may require additional approvals to install and run. We will be adding this capability in a future release