Terraform MCP server overview

This topic provides an overview of the Terraform model context protocol (MCP) server, which helps you write Terraform configuration using an AI or LLM. The server requests information directly from the Terraform registry so that your model can formulate responses using provider code and documentation as the source of truth.

Note

This feature is currently in beta. Do not use beta functionality in production environments.

Introduction

MCP standardizes how AI models can discover and interact with external tools, applications, and data sources. You can configure MCP clients, such as an AI or LLM chat, to send requests through an MCP server so that the model can provide answers to prompts that it may not have been trained on. Refer to the MCP documentation for details about how MCP works.

You can configure your AI model to connect to the Terraform MCP server so that it can request information about providers, modules, and other objects from data stored in the Terraform registry. When the Terraform MCP is available to your AI client, the model retrieves information from the most up-to-date source code stored in the registry, instead of relying on outdated training data. As a result, you can get more accurate and actionable information.

Note that to use the server, your questions must pertain to provider configuration from the registry. The LLM uses other sources, such as content from the internet, to respond to general questions about Terraform configuration or requests to generate code.

Workflow

The Terraform MCP server runs in a Docker container that you can deploy to either your local workstation or a remote instance. When running locally, start the server in stdio mode. When running on a remote instance, start the server in http mode so that you can communicate with the server using the Streamable HTTP transport. Refer to Deploy server for more information.

After deploying the server, you prompt your model to invoke MCP server tools that retrieve different types of information from the Terraform registry. Refer to Prompt your model for more information.

Terraform MCP server tools

The following table describes tools available in the Terraform MCP server:

Toolset	Tool	Description
`providers`	`resolveProviderDocID`	Queries the Terraform Registry for available documentation about a specific provider. This tool uses `serviceSlug` to return a list of provider document IDs with their titles and categories for resources, data sources, functions, or guides.
`providers`	`getProviderDocs`	Fetches the complete documentation content for a specific provider resource, data source, or function using a document ID obtained from the `resolveProviderDocID` tool. Returns the raw documentation in Markdown format.
`modules`	`searchModules`	Searches the Terraform Registry for modules based on the specified `moduleQuery` with pagination. Returns a list of module IDs with their names, descriptions, download counts, verification status, and publish dates.
`modules`	`moduleDetails`	Retrieves detailed documentation for a module using a module ID obtained from the `searchModules` tool, including inputs, outputs, configuration, submodules, and examples.
`policies`	`searchPolicies`	Queries the Terraform Registry for the appropriate Sentinel policy based on the provided query `policyQuery`. This tool uses `terraformPolicyIDs` to return a list of policies, including their name, title, and download counts.
`policies`	`policyDetails`	Retrieves detailed documentation for a policy set using a `terraformPolicyID` obtained from the `searchPolicies` tool including policy readme and implementation details.

Release notes

For information about new releases, refer to the Terraform MCP server repository.

Provide feedback

We want to hear from you. Complete the Terraform MCP server feedback form to help us improve this feature.