Share via

Dataverse MCP server reference

Important

You need to be part of the Frontier preview program to get early access to Microsoft Agent 365. Frontier connects you directly with Microsoft’s latest AI innovations. Frontier previews are subject to the existing preview terms of your customer agreements. As these features are still in development, their availability and capabilities may change over time.

Overview

Display name Description
Microsoft Dataverse MCP Server MCP Server containing tools to work with Microsoft Dataverse.

Available tools

Tool Description
create_record Enables users to insert a new record into a specified Dataverse table. Upon successful creation, the tool returns the globally unique identifier (GUID) of the newly added row, facilitating seamless integration with downstream operations and workflows.
describe_table Retrieves the complete T-SQL schema for a given Dataverse table, including field names, data types, and relationships. This provides developers with a clear understanding of the table's structure for building queries and managing data models effectively.
list_tables Returns a comprehensive list of all available tables within the current Dataverse environment. This helps users quickly discover and navigate the data landscape within their organizational environment.
read_query Executes SELECT statements to query Dataverse data directly, returning results based on specified filters and conditions. Ideal for data exploration, analytics, and integration scenarios that require dynamic data retrieval.
update_record Allows users to modify an existing record within a Dataverse table by specifying the record's unique identifier (GUID) and the fields to be updated. This ensures efficient maintenance and synchronization of data.
create_table Enables the creation of new Dataverse tables by defining the schema, including column names, data types, and relationships. This tool supports flexible schema management aligned with evolving business needs.
update_table Facilitates the modification of an existing table's schema or metadata—such as adding new columns, renaming fields, or updating constraints—without disrupting existing data or workflows.
delete_table Permanently removes a table from the Dataverse environment, including its associated schema and data. This operation is restricted to users with appropriate permissions to ensure governance compliance.
delete_record Deletes a specific record from a Dataverse table using its unique identifier. This operation supports efficient data lifecycle management and compliance with retention policies.
search Enables keyword-based search across Dataverse, allowing users to locate specific records, entities, or fields efficiently. This tool is especially useful for exploratory analysis and entity discovery within large datasets.
fetch Retrieves the complete record details from Dataverse using the entity name and record ID. This tool provides full visibility into a record's data, supporting auditing, reporting, and AI-driven insights.

Key features

Data creation and ingestion

Populate Dataverse tables with new data for business records, configurations, or operational entities.

Tools:

  • create_record – Inserts new rows into Dataverse tables and returns the record's GUID, enabling downstream automation or linking.
  • create_table – Allows developers or administrators to define new tables with a specified schema to store business data.

Data exploration and discovery

Understand the structure of existing data models and identify available entities within Dataverse.

Tools:

  • list_tables – Lists all tables available in the connected Dataverse environment.
  • describe_table – Retrieves the T-SQL schema of any selected table, including columns, types, and relationships.

Data querying and analysis

Retrieve, analyze, or validate business data for Copilot reasoning or developer workflows.

Tools:

  • read_query – Executes SELECT queries to fetch filtered data from Dataverse tables.
  • search – Searches across Dataverse using keywords to locate records that match specific criteria.
  • fetch – Retrieves complete details of a record using entity name and ID for deeper analysis.

Data updates and maintenance

Modify existing data or update schema to align with changing business logic or processes.

Tools:

  • update_record – Updates fields in an existing record, such as status, owner, or priority.
  • update_table – Adjusts a table's schema or metadata—for example, adding a new column or renaming a field.

Data deletion and lifecycle management

Manage retention policies or clean up obsolete records and tables while maintaining data governance.
Tools:

  • delete_record – Removes specific records by GUID, supporting automated cleanup or GDPR-compliant deletions.
  • delete_table – Deletes unused or deprecated tables safely from Dataverse.

Intelligent copilot and AI scenarios

Empower AI copilots and agents (e.g., Microsoft Copilot Studio, Claude Desktop) to reason over Dataverse data securely.
Tools:

  • search, read_query, and fetch – Enable copilots to dynamically locate, interpret, and respond using enterprise data.
  • create_record and update_record – Allow copilots to log new actions or modify existing records as part of conversational workflows.
  • Last updated on