Qlik MCP tools

The Qlik MCP server includes tools that let your LLM perform actions for you in your Qlik Cloud tenant. You can customize your experience by selecting which Qlik MCP tools you want available.

Access permissions and Qlik MCP tools

Access to MCP tools is controlled by your permissions. You require the Qlik MCP: Allowed permission in a custom role to access Qlik MCP tools from your LLM client.

Additionally, as the Qlik MCP tool is authorized as the connecting user, it uses your existing user role and space permissions. You also need to have permission in Qlik Cloud to use the capabilities or content type used by the tools. For example, if you do not have permission to access lineage in Qlik Cloud, you cannot use the Qlik MCP tool Qlik get lineage. Depending on your permissions, some tools may respond with no access if you cannot use the capability or feature in Qlik Cloud.

Tool availability can also be limited by your tenant's licensing.

Selecting Qlik MCP tools

The Qlik MCP tools are a catalog of purpose-built actions that an LLM can call to perform actions in your Qlik Cloud tenant. These actions include:

• Finding applications and datasets

• Inspecting fields and chart metadata

• Applying and clearing filters

• Creating sheets and charts

• Managing governed assets like glossary terms and data products.

Each tool represents a specific capability with well-defined inputs and outputs.

Available Qlik MCP Tools

The table below shows an overview of current Qlik MCP tools categories and tool names.

What the tools can each accomplish

App Discovery and Metadata

Use these tools to find the right Qlik analytics application and quickly understand what’s inside it before you build charts or apply filters. This category supports an end-to-end discovery flow: search across Qlik resources to locate relevant applications.

The following tools are available:

• Qlik_search: Search for Qlik resources (applications, datasets, data products, glossaries) by name or content.

• Qlik_describe_apps: Get comprehensive metadata for an application including fields, owner, publishing status.

• Qlik_get_fields: List all data fields available in an application for use as dimensions.

• Qlik_list_sheets: List all sheets in an app.

• Qlik_get_sheet_details: Get details about a specific sheet including all charts and their types.

• Qlik_search_spaces: Search for spaces based on a query string.

• Qlik_semantic_search_app: Perform semantic search for relevant assets in Qlik app.

Prompt example:

"I need to analyze customer churn." This simple prompt allows to:

• Use qlik_search to find applications related to “churn” or “customer retention”.

• For the best match, run qlik_describe_app to confirm it’s the correct application and see its metadata (owner and publishing status).

• Use qlik_get_fields to list available fields and identify candidate dimensions and measures like Customer, ChurnFlag, SubscriptionType, Region, and ChurnDate.

• Use qlik_list_sheets to see what dashboards already exist.

• For any promising sheet for instance, “Churn Overview,” call qlik_get_sheet_details and summarize what charts are already available and what gaps remain.

Business Glossary

This set of tools lets a user set up and maintain a governed business glossary end-to-end:

• Create a glossary

• Organize it with categories

• Add and curate terms (including editing, removing, and managing life-cycle status such as draft -verified-deprecated)

• Connect those terms to real analytics assets (applications, datasets, fields, master items).

The following tools are available:

• Qlik_create_glossary: Create a new business glossary

• Qlik_get_full_glossary_export: Get complete glossary with all terms, categories, links (costly operation)

• Qlik_get_glossary_categories: Retrieve all categories for a glossary

• Qlik_create_glossary_category: Create a new category within a glossary

• Qlik_search_glossary_terms: Search for terms within a glossary

• Qlik_get_glossary_term: Get a specific term from a glossary

• Qlik_create_glossary_term: Create a new glossary term with description, tags, relationships

• Qlik_update_glossary_term: Modify an existing glossary term

• Qlik_delete_glossary_term: Remove a glossary term

• Qlik_update_term_status: Update term status (draft, verified, deprecated). Only a steward can verify a term. Once the term is verified only a steward can modify the term

  Information note

  This operation has three statuses:

  • draft: The term is in draft state.

  • verified: The term has been verified by a steward.

  • deprecated: The term is deprecated and should not be used.

  The status names are case-sensitive and must be provided exactly as above.

• Qlik_get_glossary_term_links: Get resources linked to a glossary term

  This function supports two modes:

  • Single link mode: Provide individual parameters for instance resource_id or resource_type.

  • Batch mode: Provide a list of link dictionaries via the 'links' parameter

  Information note

  When linking to a subresource, all three subresource fields (subResourceId, subResourceName, and subResourceType) must be provided together for each link.

• Qlik_create_glossary_term_links: Link a term to apps, datasets, fields, master items

Prompt example:

"Help me build and govern a Business Glossary for our Sales domain. Create a glossary called ‘Sales Glossary’, add categories ‘Revenue’, ‘Customers’, and ‘Pipeline’. Then create terms for ‘Annual Recurring Revenue (ARR)’ and ‘Customer Churn’ with clear definitions and tags for instance finance, and sales-ops, set them to draft, and link each term to the relevant dataset fields and the KPI master items."

Datasets and Data Quality

These tools help a user understand, validate, and govern datasets by inspecting what the dataset is (metadata and data product memberships), what it contains (schema), what the data looks like (profiling statistics-distributions and a quick row sample). The tools also support stewardship workflows: improving dataset documentation (name-description) and triggering and monitoring data quality computations to refresh quality metrics over time.

The following tools are available:

• Qlik_get_dataset: Load metadata of a dataset including trust score

• Qlik_get_dataset_schema: Load the schema (column definitions) of a dataset

• Qlik_get_dataset_profile: Load profile data (statistics, distributions) of a dataset

• Qlik_get_dataset_sample: Load first 10 rows of a dataset for preview

• Qlik_get_dataset_freshness: Get last updated timestamp of a dataset

• Qlik_get_dataset_trust_score: Get the trust score of a dataset

• Qlik_get_dataset_memberships: Get data product memberships for a dataset

• Qlik_update_dataset_metadata: Update name and description of a dataset

• Qlik_update_dataset_quality: Request a data quality computation for a dataset

• Qlik_update_dataset_quality: Request a data quality computation for a dataset

• Qlik_get_dataset_quality_computation_status: Check status of a quality computation

Prompt example:

“Assess the readiness of our Customer Orders dataset for a new dashboard. Show me its metadata and current trust score, confirm the last refresh time, and list which data products it belongs to. Then load the schema and a 10-row sample to sanity-check key fields like order_date, customer_id, and net_amount. Run a dataset profile to spot missing values and outliers, and if quality metrics look stale, trigger a data quality computation and keep checking the job status until it completes. Finally, update the dataset description to document known constraints and recommended use.”

Data Exploration & Analysis

These tools support quick and ad-hoc investigation of data and existing analytics content without requiring the user to permanently build new visualizations. You can create temporary calculations, or query objects to answer “what-if” questions, inspect fields by listing distinct values or search for specific values.

The following tools are available:

• Qlik_create_data_object: Create temporary calculation objects for ad-hoc analytics queries

  Information note

  Use get_field_values() or search_field_values() before applying selections to verify values exist.

  Warning note

  Qlik performs ALL calculations, therefore:

  • Never aggregate, sum, average, or compute on returned data: values are final.

  • For different calculations, call tool again with new expressions.

  • Always apply appropriate filters-selections to limit data size and improve performance.

• Qlik_get_field_values: Get distinct values for a specific field (use before filtering)

  Information note

  For high cardinality fields, always use qlik_search_field_values() instead

  Warning note

  Use this tool or qlik_search_field_values() BEFORE creating selections-filters to verify that the values exist. This helps avoid errors when applying filters with non-existent values.

• Qlik_search_field_values: Search for specific values across fields (verify before set analysis)

  Warning note

  Before creating data objects with set analysis or selections: Use qlik_search_field_values to verify values exist. This prevents errors from non-existent filter values, especially important for: years, dates, currency codes, product names.

  Best practice example workflow:

  1. qlik_search_field_values(fieldName="payment_year", searchTerms=["2022"])

  2. Verify "2022" exists in results

  3. Create data object using {payment_year={2022}.

• Qlik_get_chart_data: Retrieve paginated data from an existing chart visualization

• Qlik_get_chart_info: Get metadata about a chart without retrieving its data

Prompt example:

“Help me investigate why North region revenue dropped last month. First, check the existing ‘Revenue by Region’ chart: show me its metadata (dimensions, measures, filters, row count) and then pull the chart data for the last two months. Before I apply any selections, list the distinct values for Region and search the field values to confirm whether ‘North’, ‘NORTH’, or ‘Northern’ is used. Then create a temporary calculation to compare month-over-month revenue and margin for North vs other regions, and highlight which product categories contributed most to the change.”

Data Products

These tools manage the full lifecycle of a data product as a governed, shareable package of datasets:

1. Create it.

2. Inspect its metadata and documentation.

3. Maintain its definition (name, description and which datasets it contains)

4. Control where and whether it’s available by moving it between spaces and activating or deactivating it.

They also support end-of-life cleanup by deleting a data product when it’s no longer needed.

The following tools are available:

• Qlik_create_data_product: Create a new data product

• Qlik_get_data_product: Get metadata for a specific data product

• Qlik_get_data_product_documentation: Get markdown documentation of a data product

• Qlik_update_data_product: Update properties (name, description, datasets) of a data product

• Qlik_update_data_product_space: Move a data product to a different space

• Qlik_update_activate_data_product: Activate a data product in a specific space

• Qlik_update_deactivate_data_product: Deactivate a data product

• Qlik_delete_data_product: Delete a data product

Prompt example:

“Create a data product called ‘Sales Analytics – Curated’ with a clear description and include these datasets: Orders, Customers, and Products. Then show me the data product metadata and pull its markdown documentation so I can review what consumers will see. Update the description to add usage guidance and add the Returns dataset as well. Move the data product to our Shared ‘Analytics’ space, activate it there for broader access, and if we later replace it with a new version, deactivate the old one. Finally, if the product is fully retired and no longer referenced, delete it.”

Lineage

This tool lets a user trace where data comes from and how it flows by retrieving upstream lineage for a dataset or app. Because each call returns only one step back, you typically repeat it recursively to build the full chain. This is useful for:

• Impact analysis.

• Troubleshooting unexpected numbers.

• Governance or auditing.

• Identifying the true sources feeding a report or dataset.

The following tools are available:

• Qlik_get_lineage: Load lineage history of a dataset or app (call recursively for full chain)

Prompt example:

“Show me the full upstream lineage for the Customer Orders dataset. Start from the dataset and walk back recursively until you reach the original source systems. For each step, summarize what the immediate parent is and note any key transformations or intermediate datasets or apps involved. Then tell me which upstream source is most likely to affect the net_amount field if it changes.”

Master Items (Dimensions & Measures)

These tools help users standardize and reuse key business logic in an app by working with master (library) dimensions and measures. You can inventory what’s already available (to avoid duplicates and encourage consistency), and you can create new reusable dimensions and measures so charts across multiple sheets use the same field definitions and calculation expressions, improving governance, maintainability, and metric alignment.

The following tools are available:

• Qlik_list_dimensionst: List all library dimensions available in the app

• Qlik_create_dimension: Create a reusable library dimension

• Qlik_update_dimension: Update an existing library dimension in a Qlik app

• Qlik_delete_dimension: Delete a library dimension from a Qlik app

• Qlik_list_measures: List all library measures available in the app

• Qlik_create_measure: Create a reusable library measure with expression

• Qlik_update_measure: Update an existing library measure in a Qlik app

• Qlik_delete_measure: Delete a library measure from a Qlik app

Prompt example:

“Help me standardize metrics in our Sales Performance app. First, list the existing library dimensions and measures so we don’t duplicate anything. Then create a new library dimension ‘Customer Segment’ based on the appropriate segment field, and create a reusable measure ‘Gross Margin %’ using our standard definition (gross margin divided by revenue, formatted as a percentage). After creating them, tell me the exact master item names I should reference when building charts so all dashboards use the same logic.”

Selections & Filtering

These tools control the interactive filter state of a Qlik app:

• Apply selections to one or more fields (using exact values or pattern-predicate-style matching).

• Inspect what filters are currently active.

• Clear selections either globally or for a specific field.

Together they support guided analysis flows, reproducible investigation steps.

The following tools are available:

• Qlik_select_values: Apply selections (filters) to fields - supports exact values and pattern matching

  Information note

  When to use selections over set analysis:

  • When you want to filter the entire app or session for multiple subsequent operations, use select_values().

  • When you need a one-time filter for a specific calculation, use set analysis in expressions

  Selections persist across all operations until cleared they affect ALL subsequent data retrievals. As best practice, for single analytical queries, prefer set analysis over selections to avoid state management overhead.

  Warning note

  • Use qlik_get_field_values() or qlik_search_field_values() first to verify that the values you want to select exist. Selecting non-existent values will fail silently.

  • The returned selections are the actual current applied selection are truth. Any fields where selection failed for instance: where values did not exist will not appear in the returned list.

• Qlik_clear_selections: Clear selections - all or specific field

  Warning noteThe returned selections are the actual current applied selection and they are the truth. Any fields where selection failed for instance, values did not exist, will not appear in the returned list.

• Qlik_get_current_selections: Get currently active selections or filters in the app

• Qlik_update_filter: Update an existing filter panel in a Qlik app

Prompt example:

“Set up my analysis context for the Executive Sales app: select Year = 2025 and Region = EMEA, and apply a pattern-based selection on Product Category to include only categories starting with ‘Cloud’. Then show me the current selections so I can confirm what’s active. After I review the KPIs, clear only the Product Category selection (keep Year and Region), and finally clear all selections to reset the app back to an unfiltered state.”

Visualization & Sheets

These tools let users assemble dashboards in a Qlik application:

• Create a new sheet as a canvas.

• Add visualizations (charts, tables, KPIs.) configured with dimensions, measures and display options.

• Place filter panels on the sheet so users can interactively slice the analysis.

Together, they cover the basic workflow of building a usable, self-service dashboard page.

The following tools are available:

• Qlik_create_sheet: Create a new empty sheet (dashboard) in the app

• Qlik_add_chart: Add a visualization (bar, line, pie, table, KPI, etc.) to a sheet

  Information note

  • Plan your query structure before making tool calls.

  • Test date-value existence with qlik_search_field_values or qlik_get_field_values first

  • Use set analysis over app-level selections for one-off queries

• Qlik_add_filter: Add a filter panel to a sheet for user-driven filtering

• Qlik_delete_object: Delete an object from a Qlik app (chart, sheet, filter, etc.)

• Qlik_show_chart: Show a rendered visualization as a Qlik snapshot

• Qlik_update_chart: Update an existing chart visualization in a Qlik app

Prompt example:

“Create a new sheet called ‘Sales Overview’ in our analytics app. Add a KPI for Total Revenue and Total Orders, a line chart showing Revenue by Month, and a bar chart showing Revenue by Region (sorted descending). Then add a filter panel with Year, Region, and Product Category so users can adjust the view. Make sure the visuals are titled clearly and the sheet is ready for an executive audience.”