Official

Power BI Modeling MCP Server

Manage and modify Power BI semantic models using natural language and AI agents.

Data last scanned today · Reviewed yesterday

Overview

This server enables AI models to interact with Power BI semantic models through the Model Context Protocol. It supports complex operations including the creation and update of tables, columns, measures, and relationships within both Power BI Desktop and Fabric environments. Developers can use it to perform bulk refactoring, rename objects, and apply modeling best practices across large semantic models. Beyond basic modeling, the tool facilitates agentic workflows by supporting TMDL and Power BI Project files. It allows AI assistants to validate DAX queries and autonomously execute structured modeling tasks. All operations are designed to handle batch processing, turning repetitive model maintenance into automated workflows.

Our verdict

This server carries a high-risk profile with a security score of 25/100. The primary risk is the complete lack of an authentication method, meaning anyone with access to the server configuration can issue modeling commands. Furthermore, the server operates with full write access without a read-only mode, necessitating extreme caution when connecting to production models. It is recommended for development environments only, and users must maintain consistent model backups to mitigate the risk of unintended AI-driven changes.

Supports batch modeling operations at scale
Integrates with TMDL and Power BI Project files
Zero direct dependencies reduce supply-chain surface
Includes built-in DAX query validation capabilities

No authentication method for server access
Lacks a read-only mode for safe exploration
Community-maintained rather than official vendor software

Tools

Tool	Description	Risk
connection_operations	Connect to Power BI Desktop or Fabric workspaces	medium
database_operations	Manage semantic models - connect, create, update, list databases, import/export TMDL folders, deploy to Fabric	high
transaction_operations	Control database transactions (begin, commit, rollback, get status)	medium
model_operations	Work with the overall model (get, create, update, refresh, get stats, rename)	high
table_operations	Manage tables (create, update, delete, get, list, refresh, rename)	high
column_operations	Manage individual table columns (create, update, delete, get, list, rename)	high
measure_operations	Manage individual DAX measures (create, update, delete, get, list, rename, move between tables)	high
relationship_operations	Handle relationships between tables (create, update, delete, activate/deactivate, find)	high
dax_query_operations	Execute, validate, and generate DAX queries against the model	low
trace_operations	Perform trace operations on semantic model to capture and analyze Analysis Services events.	low
partition_operations	Manage table partitions (create, update, delete, refresh specific partitions)	high
user_hierarchy_operations	Work with user-defined hierarchies (create, update, delete levels, reorder)	high
calculation_group_operations	Manage calculation groups and calculation items for time intelligence and other calculations	high
security_role_operations	Configure security roles and row-level security (RLS) table permissions	high
perspective_operations	Manage perspectives and their members (filtered views of the model for different audiences)	medium
named_expression_operations	Work with named expressions and Power Query parameters (create, update, delete, get, list, rename)	medium
function_operations	Manage individual DAX user-defined functions	medium
culture_operations	Manage cultures for multi-language support (create, update, delete, get valid culture names)	medium
object_translation_operations	Handle translations for model objects across different cultures/languages	medium
calendar_operations	Manage calendar objects and time intelligence column groups	medium
query_group_operations	Organize and manage query groups for Power Query expressions	medium

Compatibility

Client	Local	Docker	Remote	Read-only
ChatGPT
Claude Desktop
Cursor
VS Code
Windsurf

Frequently asked questions

›What types of Power BI metadata can this server modify?

This server can only execute modeling operations such as managing tables, columns, measures, and relationships. It cannot modify other metadata types like report pages or diagram layouts.

›Is it possible to use this server outside of Visual Studio Code?

Yes, it can be configured in any MCP-compliant client using NPX with Node.js or by manually downloading and executing the server binary.

›What are the recommended AI models for best performance?

Deep-reasoning models like GPT-5 or Claude Sonnet 4.5 are recommended to ensure higher quality and relevance of the modeling results.

›Are there any specific risks when using this server with AI agents?

AI agents may produce unexpected or inaccurate results that could lead to unintended changes, and they might unintentionally expose sensitive model information. You should always create a backup of your model before performing any operations.

›Why might the server not appear in GitHub Copilot even after installation?

Ensure that the 'MCP servers in Copilot' option is enabled in your GitHub.com account settings, as this is disabled by default for enterprise accounts.