Collibra MCP tools

Updated: October 10, 2026

The Collibra MCP server exposes tools in three categories. The full list, including source code and schema definitions, is in the Git repository.

Note Most tools work with the default scopes granted to any authenticated Collibra user. Tools that require additional scopes are marked in the tables below. If a tool returns a permission error, you are missing the required scope. Contact your Collibra administrator to request access.

Data quality tools with data-quality in the Capability flag column are off by default. For more information about how to turn them on, go to Data quality tools.

Read tools

Read tools return data from Collibra without making any changes.

Tool	Description	Capability flag	Extra scopes required	Uses Collibra Units?
`discover_business_glossary`	Ask questions about business terms and definitions using natural language.	N/A	`dgc.ai-copilot`	Yes
`discover_data_assets`	Query available data assets using natural language.	N/A	`dgc.ai-copilot`	Yes
`dq_get_job`	Retrieve the full definition of a data quality job by `name`, including the type (PUSHDOWN or PULLUP), Edge site, connection, schema and table, source SQL, run-date window, configured monitors (adaptive and custom DQ rules), notifications, and schedule. If no job name matches exactly, the tool returns the jobs whose name contains the text as candidates.	`data-quality`	N/A	No
`dq_get_job_run`	Retrieve the full details of a data quality job run by `run_id`, including the lifecycle status, activity, and timing. When the run is in a terminal state (FINISHED, CANCELLED, or FAILED), the details also include the overall score, row count, execution time, and the score breakdown per monitor (adaptive and custom DQ rules). Fields that apply only to finished runs are not shown while the run is in progress.	`data-quality`	N/A	No
`dq_search_job_runs`	Search the run history of data quality jobs. Filter by `job_name` (partial match), `status`, or job_type (PUSHDOWN or PULLUP). If you do not use a filter, the tool returns all runs that you can access. Returns the run ID, job name, status, and start time of each run. Paginated with `page` and `page_size`.	`data-quality`	N/A	No
`dq_search_jobs`	Search data quality job definitions. Filter by `name` (partial match), `schema`, `table`, `edge_site`, or `job_type` (PUSHDOWN or PULLUP). If you do not use a filter, the tool returns all jobs that you can access. Returns the name, type, schema, table, and Edge site of each job. Job definitions do not have a status. To get the status of runs, use `dq_search_job_runs`. Paginated with `page` and `page_size`.	`data-quality`	N/A	No
`find_data_quality_rules`	Search existing data quality rules (monitors) across jobs. Filter by exact `jobName`, `columnName`, or part of the rule name. Combine `jobName` and `columnName` to check whether a rule already exists on a target column before creating a new one. Returns job, column, type, status, and SQL of each rule. Paginated.	`data-quality`	N/A	No
`generate_data_quality_rule_sql`	Uses SQL AI to convert a plain-language description of a check into rule SQL, so you can you to create rules without manually writing SQL. Always validate the generated SQL using `validate_data_quality_rule` before applying it. Requires `edgeSiteId` and `connectionId`, both available from `create_data_quality_job`.	`data-quality`	N/A	Yes
`get_asset_details`	Retrieve detailed information about a specific asset by UUID. This includes the asset's assignable attribute schema, which lists all attributes the asset can hold, including empty ones.	N/A	N/A	No
`get_business_term_data`	Trace a business term back to its connected physical data assets.	N/A	N/A	No
`get_column_semantics`	Retrieve data attributes, measures, and business assets connected to a column.	N/A	N/A	No
`get_data_quality_rule`	Retrieve the definition of a data quality rule (monitor) on a job, including its type, SQL, filter, tolerance, and active or suppressed state.	`data-quality`	N/A	No
`get_data_quality_rule_results`	Retrieve the results of a data quality rule for each job run, including the score, breaking and passing record counts, pass or fail status, and any exception. Returns the newest results first by default. Paginated with `offset` and `limit`.	`data-quality`	N/A	No
`get_data_quality_rule_template`	Retrieve a data quality rule template by `ruleTemplateName`, including its parameterized SQL, dialect, dimensions, default tolerance, built-in flag (`isSystem`), and the number of rules deployed from it.	`data-quality`	N/A	No
`get_lineage_downstream`	Get downstream technical lineage (consumers) for a data entity.	N/A	N/A	No
`get_lineage_entity`	Get metadata about a specific entity in the technical lineage graph.	N/A	N/A	No
`get_lineage_transformation`	Get details and logic of a specific data transformation.	N/A	N/A	No
`get_lineage_upstream`	Get upstream technical lineage (sources) for a data entity.	N/A	N/A	No
`get_measure_data`	Trace a measure back to its underlying physical columns and tables.	N/A	N/A	No
`get_table_semantics`	Retrieve the semantic layer for a table: columns, data attributes, and connected measures.	N/A	N/A	No
`list_asset_types`	List available asset types.	N/A	N/A	No
`list_data_contract`	List data contracts with pagination.	N/A	N/A	No
`list_data_quality_rule_templates`	List the data quality rule templates (built-in and custom) that are available in the connected environment. Each template is a parameterized SQL pattern that you can deploy with `deploy_data_quality_rule_template`. Filter by `name`, `dimension`, or `isSystem` (built-in or custom). Paginated.	`data-quality`	N/A	No
`prepare_create_asset`	Enumerate available asset types and domains and hydrate the attribute and relation schema for a chosen pair. Use before `create_asset`.	N/A	N/A	No
`pull_data_contract_manifest`	Download the manifest for a data contract.	N/A	N/A	No
`search_asset_keyword`	Wildcard keyword search for assets.	N/A	N/A	No
`search_data_class`	Search for data classes with filters.	N/A	`dgc.data-classes-read`	No
`search_data_classification_match`	Search for associations between data classes and assets.	N/A	`dgc.classify`, `dgc.catalog`	No
`search_lineage_entities`	Search for entities in the technical lineage graph.	N/A	N/A	No
`search_lineage_transformations`	Search for transformations in the technical lineage graph.	N/A	N/A	No
`validate_data_quality_rule`	Validate the SQL or definition of a data quality rule against the source before you save or run the rule. Returns if the rule is valid and the message from the engine. Requires `edgeSiteId`, `connectionId`, and `schemaName`, which identify the data quality connection that the rule runs against. To discover connections, use `create_data_quality_job`.	`data-quality`	N/A	No

Write tools

Write tools create, modify, or delete data in Collibra.

Tool	Description	Capability flag	Extra scopes required	Uses Collibra Units?
`add_data_classification_match`	Associate a data class with an asset.	N/A	`dgc.classify`, `dgc.catalog`	No
`create_asset`	Create a new asset of any type. Resolves asset type, domain, status, and attributes by name or UUID.	N/A	N/A	No
`create_data_quality_job`	Set up automated data quality monitoring on a database table and start the first run. A data quality job profiles a table and monitors it over time for problems, such as sudden drops in row count, increases in null or empty values, duplicates, and other anomalies. The tool includes discovery, preview, and creation. For discovery, you specify one field at a time (connection, data source, schema, and table), or you specify a catalog Table asset. Optional settings: selected columns, monitors and adaptive baseline settings, sample size, row filter, time slice, recurring schedule, back-runs, and notifications. `jobType`: Choose `PUSHDOWN` to run checks in the source database, or `PULLUP` to pull data into Spark. Confirm checkpoint: When `confirm=false` (default), the tool returns the exact request and does not create the job. Set `confirm=true` to create and run the job.	`data-quality`	N/A	No
`create_data_quality_rule`	Create a data quality rule (monitor) on an existing DQ job. `monitorType`: Choose `FREEFORM_SQL` for a full SQL query or `SIMPLE_SQL` for a single-column check. The rule is active and not suppressed by default. Confirm checkpoint: When `confirm=false` (default), the tool returns a preview of the rule and SQL and does not create the rule. Set `confirm=true` to create the rule. Requires permission to create rules on the target job.	`data-quality`	N/A	No
deploy_data_quality_rule_template	Deploy a rule template as rules on one or more job and column targets in bulk. The DQ service resolves dialect-specific SQL and names each rule `{templateName}_{columnName}`. Confirm checkpoint: When `confirm=false` (default), the tool previews the template and targets and does not deploy. Set `confirm=true` to deploy. Requires permission to deploy templates and create rules on the target jobs.	`data-quality`	N/A	No
`dq_cancel_job_run`	Cancel a data quality job run that is in progress. Specify the `jobRunId` or the `jobName`, not both. With `jobRunId`, the tool does not cancel a run that is already in a terminal state (FINISHED, FAILED, or CANCELLED). With `jobName`, the tool finds the runs that you can cancel. If there is only one, the tool cancels it. If there are more, the tool returns them as candidates. Run the tool again with the `jobRunId` of the run to cancel. There is no confirm checkpoint. The state check is the safety mechanism. You cannot undo a cancellation.	`data-quality`	N/A	No
`dq_delete_job`	Permanently delete a data quality job definition by `jobName`, with all of its runs, rules, monitors, and results. You cannot undo this. If a run is in progress, the service can refuse the delete. Cancel the run first with `dq_cancel_job_run`. To delete one run, use `dq_delete_job_run`. To edit the configuration of a job, use `dq_update_job`. Confirm checkpoint: When `confirm=false` (default), the tool does not delete the job. It returns a summary of the job: job type, Edge site, connection, schema or table, source query, and schedule. To delete the job, run the tool again with the same `jobName` and `confirm=true`.	`data-quality`	N/A	No
`dq_delete_job_run`	Permanently delete a completed data quality job run and all of its per-run results (profile, scan, monitor, rule, and alert output). You cannot undo this. Specify the `jobRunId` or the `jobName`. A `jobName` never deletes a run directly. It only returns candidate runs for review. You can delete only terminal runs. To stop a run that is in progress, use `dq_cancel_job_run` first. Confirm checkpoint: When `confirm=false` (default), the tool returns the run details and does not delete the run. To delete the run, run the tool again with the same `jobRunId` and `confirm=true`.	`data-quality`	N/A	No
`dq_update_job`	Update some fields of an existing data quality job, identified by `jobName`. Specify only the fields to edit. All other fields stay the same. You can update the scan SQL (`sourceQuery`), run-date window, recurring schedule, monitor set and adaptive baseline, notifications, PUSHDOWN compute settings, PULLUP sizing, and data location (for a moved or renamed table). To turn off a schedule, set `scheduleRepeat=NEVER`. Merge behavior: Most settings update field by field. These settings are replaced fully: `monitors` (the tool turns off monitors that you do not specify), the schedule, and the notification configuration (when you set a `notify` field). Specify again the parts that you want to keep. Data location fields apply on top of the current location. For example, you can edit only `tableName`. Limitations: You cannot edit the job type (PUSHDOWN or PULLUP). The update does not include back-runs. Column selection, row filters, sampling, and the time slice are part of `sourceQuery`, and the tool does not update them for you. To manage rules, use `create_data_quality_rule` or `deploy_data_quality_rule_template`. To remove a job, use `dq_delete_job`. Confirm checkpoint: When `confirm=false` (default), the tool returns a before-and-after comparison (`changes`) and the exact PATCH body, and does not update the job. To apply the updates, run the tool again with the same inputs and `confirm=true`.	`data-quality`	N/A	No
`edit_asset`	Edit an existing asset. Supports updating attributes, properties, relations, tags, and responsibilities. The following operations are supported: `update_attribute`, `add_attribute`, `remove_attribute`: Change append or clear an attribute value, such as `Definition` or `Note`. `update_property`: Rename the asset (`name`), change its `displayName`, or change its `statusId`. Status name and UUID are accepted. `add_relation`, `remove_relation`: Link or unlink the asset to another asset by relation role, such as `is synonym of`. `add_tag`: Append a free-text tag without replacing existing ones. `set_responsibility`: Assign a user or group to a resource role, such as `Steward` or `Owner`, by username, email, or UUID. `remove_responsibility`: Unassign a user or group from a resource role. This only applies to directly-assigned responsibilities, not inherited ones.	N/A	N/A	No
`init_data_contract`	Initialize a new data contract asset governing a Data Product Port, with an optional initial manifest.	N/A	`dgc.data-contract`	No
`push_data_contract_manifest`	Upload a manifest for a data contract.	N/A	`dgc.data-contract`	No
`remove_data_classification_match`	Remove an association between a data class and an asset.	N/A	`dgc.classify`, `dgc.catalog`, `dgc.data-classes-edit`	No

Data quality tools

Data quality tools are generally available, but they are off by default. Some of these tools write to Collibra, and dq_delete_job and dq_delete_job_run permanently delete data. For this reason, you turn on all data quality tools together with one capability flag.

To turn on data quality tools, do one of these steps:

Pass --data-quality at startup.
Set the COLLIBRA_MCP_DATA_QUALITY environment variable.
Set the mcp.data-quality field in your mcp.yaml file.

For configuration details, go to the Git repository.

When the data quality capability is off, the tools with data-quality in the Capability flag column are not registered. They do not appear in the tool list, and you cannot call them.

Note You cannot use the enabled tools list to turn on a data quality tool when the data quality capability is off. For example, if you add create_data_quality_rule to the enabled tools but do not turn on the capability, the server does not register the tool.

data-quality is not an experimental feature name. Do not use --experimental=data-quality. The server logs a warning and does not turn on the data quality tools.

If you also turn on the skills experimental feature, the server provides the data quality skill guides (collibra/dq-rules and collibra/dq-rule-workbench) only when the data quality capability is on.

Experimental tools

Experimental tools are opt-in and off by default. Collibra can update or remove them without a deprecation cycle. To turn on an experimental feature, pass --experimental=<name> at startup, set the COLLIBRA_MCP_EXPERIMENTAL environment variable, or add the feature name under mcp.experimental in your mcp.yaml file. For configuration details, go to the Git repository.

Note Experimental tools apply to the local MCP server only. The remote server’s tool set is managed by Collibra.

Feature flag	Tools added	Description
`skills`	`list_collibra_skills`, `load_collibra_skill`	Serves an embedded catalog of Markdown guides documenting multi-step Collibra workflows, such as discovery, lineage, and asset creation, for the connecting AI model.