Connect AI clients to HCP Terraform with the MCP server

The remote Terraform Model Context Protocol (MCP) server lets AI clients and AI coding agents access your HCP Terraform organizations, your private registry, and the public Terraform registry. HashiCorp hosts the server at https://mcp.terraform.io/mcp, so you do not need to install or operate an MCP server.

Note

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

Requirements

You must have the following to connect to the remote Terraform MCP server:

An HCP Terraform account on any plan, including Free, during the beta.
An HCP Terraform user API token or team API token.
An AI client that supports MCP servers over HTTP and lets you send a custom HTTP header with each request.

Note

HashiCorp Cloud Platform (HCP) Europe organizations only support the Splunk, AWS Service Catalog and Run task integrations at this time. To learn more about HCP Europe, refer to Use HCP Terraform in Europe.

Connect your AI client

Add https://mcp.terraform.io/mcp to your AI client's MCP configuration and send your API token in the Authorization: Bearer header. The server can only perform actions that your token has permission to perform.

Refer to Connect to the remote Terraform MCP server in the Terraform MCP server documentation for client configuration examples, available tools, and limitations.