Edit

Microsoft Dataverse plugin for AI coding agents (preview)

Note

  • This feature is in preview.
  • Preview features aren’t meant for production use and may have restricted functionality. These features are available before an official release so that customers can get early access and provide feedback.

Microsoft Dataverse plugin for AI coding agents are an open-source set of instructions that help supported coding agents perform Dataverse development and administration tasks. You describe the outcome you want in natural language, and the agent selects the appropriate skill and tool.

The plugin doesn't replace the Dataverse development tools. It gives the agent task-specific guidance for using the Dataverse CLI, Dataverse MCP server, Dataverse SDK for Python, Microsoft Power Platform CLI, and Dataverse Web API.

The source is available in the microsoft/Dataverse-skills GitHub repository.

Why use the Dataverse plugin?

Dataverse development frequently requires selecting and coordinating several tools. For example, creating a data model, loading records, and moving the changes to another environment can involve the Dataverse SDK for Python, the Dataverse CLI, Power Platform CLI, and the Web API.

The plugin provides the coding agent with guidance for:

  • Loading the correct skill that identifies the right tool to accomplish the task.
  • Selecting a tool based on the operation and data volume.
  • Discovering environments and configuring authentication.
  • Creating tables, columns, relationships, forms, and views in a solution.
  • Reading, analyzing, importing, and updating Dataverse data.
  • Exporting, importing, and validating solutions.
  • Performing supported environment and security administration tasks.
  • Applying confirmation steps before changes that affect an environment, solution, data, or security configuration.

This guidance reduces the amount of tool-specific syntax you need to provide in a prompt. You should still review proposed changes, confirm the target environment, and validate the result.

How the plugin works

The coding agent plugin loads an overview skill for Dataverse context and then selects one or more specialized skills based on your request. A single request can use several skills. For example, a request to create a recruiting data model, add sample data, and package it can use the metadata, data, query, and solution skills.

The agent selects among these supported tools:

Tool Typical use
Dataverse MCP server Interactive queries, schema discovery, and small record or metadata operations
Dataverse CLI Authentication, headless data operations, API discovery and invocation, and the local MCP proxy
Dataverse SDK for Python Scripted schema operations, bulk data operations, paging, transformations, and analytics
Power Platform CLI Solution application lifecycle management (ALM), environment operations, and role assignment
Dataverse Web API Operations that aren't available through a managed tool

For descriptions of all the included skills, see Microsoft Dataverse plugin for AI coding agents reference.

Prerequisites

You need:

  • A Dataverse environment available through Power Apps, Dynamics 365, or another applicable Power Platform plan. For a nonproduction development environment, you can use the Power Apps Developer Plan.
  • A user account with access to the environment and the privileges required for the operations you request.
  • A supported coding agent with plugin support: GitHub Copilot, Claude Code, Cursor, or Codex.
  • Permission to install or run any required local tools.

The dv-connect skill checks the local toolchain and guides installation or configuration when needed. Depending on the requested operations, the toolchain can include Git, Node.js, Python 3, .NET SDK, Azure CLI, Power Platform CLI, the Dataverse CLI, and the Dataverse SDK for Python.

Dataverse MCP access also requires tenant administrator consent and per-environment client allowlisting. The connection flow identifies these requirements and provides the appropriate instructions, but doesn't bypass administrator approval.

Install the plugin

Install the plugin for your coding agent.

GitHub Copilot

/plugin install dataverse@awesome-copilot

Claude Code

/plugin install dataverse@claude-plugins-official

You can also run /plugin, open Discover, and search for dataverse.

Cursor

In agent chat, enter:

/add-plugin dataverse

Alternatively, in Cursor, go to Settings > Plugins, search for Dataverse, open Microsoft Dataverse, and select Add to Cursor.

Codex

In the Codex app:

  1. Open Plugins > Add marketplace.
  2. For Source, enter https://github.com/microsoft/Dataverse-skills.git.
  3. Leave Git ref and Sparse paths empty, and then add the marketplace.
  4. Open the dataverse-skills marketplace and install the dataverse plugin.

For the Codex CLI, add the marketplace:

codex plugin marketplace add microsoft/Dataverse-skills

Then open /plugins and install dataverse.

Connect to Dataverse

After installation, ask your coding agent:

Connect to my Dataverse environment.

The dv-connect skill:

  1. Checks whether the workspace is already configured.
  2. Verifies the required tools and installs or updates them when supported.
  3. Discovers the environments available to your account and asks you to select one.
  4. Configures Dataverse CLI and Power Platform CLI authentication.
  5. Creates local workspace configuration and excludes credentials and generated artifacts from source control.
  6. Registers the Dataverse MCP server for the coding agent.
  7. Verifies the connection.

Some coding agents must be restarted after MCP registration before MCP tools become available. Follow the instructions provided by the connection flow.

To verify the connection after restarting, ask:

List the tables in my Dataverse environment.

Example prompts

Describe the task and relevant constraints. You don't normally need to identify a skill or tool.

  • "Show me my open opportunities over $100,000 that close this quarter."
  • "Import this CSV into the contacts table. Use an alternate key so I can safely run the import again."
  • "Create a customer feedback table with name, rating, and comment columns in the CustomerService solution."
  • "Create 500 ticket records efficiently in the new_ticket table. Each ticket needs a unique title and a priority value."
  • "Export the CustomerService solution from development, unpack it into the repository, and validate the components."
  • "Show me the audit settings for this environment."
  • "Assign the Basic User role to user@contoso.com in the development environment."

Include the target environment, solution, expected volume, and safety constraints when they're important. The agent should still confirm the environment before making the first change in a session.

Safety and guardrails

The skills combine Dataverse platform security with behavioral instructions for the coding agent. Dataverse enforces platform controls. Agent-level guardrails guide how the agent plans, confirms, and executes an operation, but they aren't a substitute for reviewing the proposed action.

Supported operations

Category Examples Requirement
Read Query records, inspect tables and columns, search metadata Dataverse read privileges
Create Create records, tables, columns, relationships, forms, and views Corresponding create and customization privileges
Update Update records, schema, forms, views, and supported environment settings Corresponding update or administrator privileges
Import and export Import CSV data, export or import solutions, run bulk data operations Tool-specific privileges and access to the target environment
Delete Delete records or metadata components Corresponding delete or customization privileges

Authentication and authorization

Interactive developer authentication uses Microsoft identity flows and caches tokens in the operating system credential store. Service-principal authentication is available for unattended scenarios. The skills hold tokens in memory rather than persisting them.

The coding agent execution can't exceed the permissions of the authenticated identity. Dataverse enforces table privileges, field-level security, ownership and sharing, and administrator permissions on every request. If the identity lacks a required privilege, the operation fails.

Dataverse MCP has additional authorization layers:

  1. The developer authenticates.
  2. A tenant administrator grants consent to the MCP client application.
  3. An environment administrator allows the client application in each target environment.

Other tools used by the plugin, such as the Dataverse SDK for Python and Power Platform CLI, authenticate directly and aren't subject to these MCP-specific controls.

Confirmation before changes

The plugin instructions direct the agent to:

  • Show the target environment URL and get confirmation before the first operation against that environment in a session.
  • Verify the active organization connection.
  • Ask which solution should contain new metadata.
  • Show existing publishers and get confirmation before selecting or creating a publisher prefix. A prefix is permanent for components created with it.
  • Preview security changes, environment-setting changes, bulk operations, and other stateful actions before execution.

Irreversible and high-impact operations

Some operations need extra review:

  • Bulk delete bypasses the recycle bin. The administration skill requires explicit acknowledgment that includes all records and the table logical name before it schedules an unfiltered bulk deletion.
  • Deleting a table also removes dependent columns, forms, views, and other components.
  • Importing a solution into a production environment can affect all users. The solution skill includes post-import validation guidance.
  • Bulk imports should use alternate keys and upsert when reruns are expected, and should be divided into appropriate batches.
  • Assigning roles or using administrator self-elevation changes the security posture. The security skill requires explicit preview and confirmation. Microsoft Purview logs self-elevation.

Credentials, data, logging, and telemetry

  • Interactive tokens are stored in the operating system credential store. Service-principal tokens are held in memory.
  • The connection flow excludes local configuration files that can contain credentials from source control.
  • Client-side logs can contain tool execution events, request metadata, status codes, errors, and validation failures. They aren't intended to contain record data, credentials, or personal information.
  • The environment's existing Dataverse audit and telemetry policies govern requests.
  • Requests can include application metadata that identifies the plugin version, skill, and coding-agent type. Prompts, tool arguments, record data, file contents, and command strings aren't included in this attribution metadata.

For security issues, follow the instructions in the repository SECURITY.md. For product feedback or guardrail requests, create an issue.

Related content

  • Last updated on